Ecological Studies, Vol. 131
Analysis and Synthesis
Edited by
M.M. Caldwell, Logan, USA
G. Heldmaier, Marburg, Germany
O.L. Lange, Wfuzburg, Germany
H.A. Mooney, Stanford, USA
E.-D. Schulze, Bayreuth, Germany
U. Sommer, Kiel, Germany
Ecologica! Studies
Volumes published since 1992 are listed at the end of this book.
Springer-Science+Business Media, LLC
Erik Jeppesen Martin S~ndergaard
Morten S~ndergaard Kirsten Christoffersen
Editors
The Structuring Role of
Submerged Macrophytes
in Lakes
With 117 illustrations
t
Springer
Erik Jeppesen
Department of Lake and Estuarine Ecology
National Environmental Research Institute
DK-8600 Silkeborg
Denmark
Martin S!I!ndergaard
Department of Lake and Estuarine Ecology
National Environmental Research Institute
DK-8600 Silkeborg
Denmark
Morten Sl/lndergaard
Freshwater Biological Laboratory
University of Copenhagen
DK-3400 Hillerl/ld
Denmark
Kirsten Christoffersen
Freshwater Biological Laboratory
University of Copenhagen
DK-3400 Hiller!l!d
Denmark
Cover iIIustration: As it will appear from this book, submerged macrophytes may have an
important impact on the nutrient dynamics, trophic structure, and trophic interactions of shallow
lakes. Within certain nutrient limits, submerged macrophytes may via a number of feedback
mechanisms maintain a clearwater state despite increased nutrient supply. Drawing by Bjl'lm
Bachmann and Erik Jeppesen.
Library of Congress Cataloging-in-Publication Data
The structuring role of submerged macrophytes in lakeslErik Jeppesen
... [et al.].
p. cm.-{Ecological studies; v. 131)
lncludes bibliographical references and index.
ISBN 978-1-4612-6871-0
ISBN 978-1-4612-0695-8 (eBook)
DOI 10.1007/978-1-4612-0695-8
1. Lake et:ology. 2. Lake plants-Ecology. I. Jeppesen, Erik.
II. Series.
QH541.5.L3S77 1997
577.63-dc21
97-22884
Printed on acid-free paper.
© 1998 Springer Science+Business Media New York
Originally published by Springer-Verlag New York Berlin Heidelberg in 1998
Softcover reprint ofthe hardcover Ist edition 1998
Ali rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer Science+Business Media, LLC), except for brief excerpts in
connection with reviews or scholarly analysis. Use in connection with any form of information storage
and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed is forbidden.
The use of general descriptive names, trade names, trademarks, etc., in this publication, even if the former
are not especially identified, is not to be taken as a sign that such names, as understood by the Trade
Marks and Merchandise Marks Act, may accordingly be used freely by anyone.
Production coordinated by Princeton Editorial Associates and managed by Francine McNeill;
manufacturing supervised by Jeffrey Taub.
Typeset by Princeton Editorial Associates, Princeton, NJ.
987 654 3 2 I
ISBN 978-1-4612-6871-0
ISSN 0070-8356
SPIN 10632906
Preface
Submerged macrophytes have been the object of intensive research, and a large
body of literature exists on their growth, reproduction, and physiology. Several
studies have focused on the interactions between submerged macrophytes and
other autotrophic components and the impact of the plants on the dynamics of
nutrients, dissolved organic and inorganic carbon, oxygen, and pH. Comparatively
few studies have dealt with the ability of submerged macrophytes to modulate the
structure and dynamics of pelagic and benthic food webs. Recently, however, the
amount of research into the structuring role of submerged macrophytes in food
webs has markedly increased, and the results obtained so far suggest that submerged macrophytes are of significant importance for the food web interactions
and environmental quality of lakes, even at relatively low areal plant coverage.
For example, plants affect the interactions between predacious, planktivorous, and
benthivorous fish and between fish and invertebrates, including key organisms
such as large zooplankton and snails. Changes in these interactions in turn may
have cascading effects on the entire food web in both the pelagial and the littoral
zone.
To provide a forum for discussion of recent results in this growing field of
research and to define future research needs, a workshop was held on 16 to 20 June,
1996, at the Freshwater Centre in Silkeborg, Denmark. The present book is a result
of the workshop. It is divided into three parts. The first part consists of 10 thematic
chapters (Chapters 1 to 10) describing how submerged macrophytes influence
various biological and biogeochemical interactions in lakes. These chapters are
v
VI
Preface
written by authors having specialized knowledge within the field treated. Cascading effects through the food web as a result of changes in resource and
predator/grazer control are given main emphasis in several of these chapters. The
authors were given the option of either writing a state-of-the-art review or discussing the subject based on mainly their own investigations. The second part consists
of 18 case studies (Chapters 11 to 28) related to the thematic chapters, and the third
part (Chapters 29 to 31) summarizes three of the workshop's cross-subject discussions. The authors were here given the option of writing a summary of the
discussion or treating the subject more extensively, using the workshop discussions as their starting point.
We thank translator Anne Mette Poulsen of the National Environmental Research Institute for her efficient help in planning and arranging the workshop and
in the subsequent editing phase. We are also grateful to assistant editor Janet
Slobodien of Springer-Verlag and the staff of Princeton Editorial Associates for
fruitful and efficient cooperation. Finally, we thank the Strategical Environmental
Research Programme, the Danish Natural Science Research Council, and the
National Environmental Research Institute for the financial support that made this
workshop possible.
The Editors
Contents
Preface
Contributors
V
Xl
Section 1. General Themes
1. Fish-Habitat Interactions Mediated via Ontogenetic Niche Shifts
Lennart Persson and Larry B. Crowder
2. Influence of Submerged Macrophytes on Trophic Interactions
Among Fish and Macroinvertebrates
Sebastian Diehl and Ryszard Komij6w
3. Complex Fish-Snail-Epiphyton Interactions and Their Effects on
Submerged Freshwater Macrophytes
Christer Bronmark and Jan Vermaat
4. Interactions Between Periphyton, Nonmolluscan Invertebrates,
and Fish in Standing Freshwaters
John Iwan Jones, Brian Moss, and Johnstone O. Young
5. Impact of Submerged Macrophytes on Fish-Zooplankton
Interactions in Lakes
Erik Jeppesen, Torben L. Lauridsen, Timo Kairesalo, and
Martin R. Perrow
3
24
47
69
91
vii
Contents
viii
6. Impact of Submerged Macrophytes on Phytoplankton in
Shallow Freshwater Lakes
Martin S0ndergaard and Brian Moss
115
7. Role of Submerged Macrophytes for the Microbial
Community and Dynamics of Dissolved Organic Carbon in
Aquatic Ecosystems
Robert G. Wetzel and Morten S0ndergaard
133
8. Impact of Herbivory on Plant Standing Crop: Comparisons
Among Biomes, Between Vascular and Nonvascular Plants,
and Among Freshwater Herbivore Taxa
David M. Lodge, Greg Cronin, Ellen van Donk, and
Adrienne J. Froelich
9. Interactions Between Grazing Birds and Macrophytes
149
175
Stuart F. Mitchell and Martin R. Perrow
10. Effects of Submerged Aquatic Macrophytes on Nutrient
Dynamics, Sedimentation, and Resuspension
John W. Barko and William F. James
197
Section 2. Case Studies
11. Macrophyte Structure and Growth of BluegiU (Lepomis
macrochirus): Design of a Multilake Experiment
Stephen R. Carpenter, Mark Olson, Paul Cunningham, Sarig Gafny,
Nathan Nibbelink, Tom Pellett, Christine Storlie, Anett Trebitz,
and Karen Wilson
12. Vertical Distribution of In-Benthos in Relation to Fish and
Floating-Leaved Macrophyte Populations
Ryszard Koruij6w and Brian Moss
13. Horizontal Migration of Zooplankton: Predator-Mediated Use
of Macrophyte Habitat
Torben L. Lauridsen, Erik Jeppesen, Martin S0ndergaard, and
David M. Lodge
14. Changing Perspectives on Food Web Interactions in Lake
Littoral Zones
Larry B. Crowder, Elizabeth W. McCollum, and Thomas H. Martin
15. Bacterioplankton and Carbon Thrnover
in a Dense Macrophyte Canopy
Morten S0ndergaard, Jon Theil-Nielsen, Kirsten Christoffersen,
Louise Schluter, Erik Jeppesen, and Martin S0ndergaard
16. Cascading Effects on Microbial Food Web Structure in a Dense
Macrophyte Bed
Klaus Jurgens and Erik Jeppesen
217
227
233
240
250
262
Contents
17. Abundance, Size, and Growth of Heterotrophic
NanoflageUates in Eutrophic Lakes with Contrasting
Daphnia and Macrophyte Densities
Kirsten Christoffersen
18. What Do Herbivore Exclusion Experiments TeU Us?
An Investigation Using Black Swans (Cygnus atratus)
and Filamentous Algae in a Shallow Lake
Robert T. Wass and Stuart F. Mitchell
19. Switches Between Clear and Turbid Water States in a
Biomanipulated Lake (1986--1996): The Role of Herbivory
on Macrophytes
Ellen Van Donk
20. Macrophyte-Waterfowl Interactions: Tracking a Variable
Resource and the Impact of Herbivory on Plant Growth
Martin S0ndergaard, Torben L. Lauridsen, Erik Jeppesen, and
Lise Bruun
21. Influence of Macrophyte Structure, Nutritive Value, and
Chemistry on the Feeding Choices of a Generalist Crayfish
Greg Cronin
22. Concordance of Phosphorus Limitation in Lakes:
Bacterioplankton, Phytoplankton, Epiphyte-Snail Consumers,
and Rooted Macrophytes
Robert E. Moeller, Robert G. Wetzel, and Craig W. Osenberg
23. Sources of Organic Carbon in the Food Webs of Two Florida
Lakes Indicated by Stable Isotopes
Mark V. Hoyer, Binhe Gu, and Claire L. Schelske
24. Importance of Physical Structures in Lakes: The Case of Lake
Kinneret and General Implications
Avital Gasith and Sarig Gafny
25. Clear Water Associated with a Dense Chara Vegetation in the
Shallow and Turbid Lake Veluwemeer, The Netherlands
Marcel S. Van den Berg, Hugo Coops, Marie-Louise Meijer, Marten
Scheffer, and Jan Simons
IX
274
282
290
298
307
318
326
331
339
26. Alternative Stable States in Shallow Lakes: What Causes a Shift?
353
Irmgard Blindow, Anders Hargeby, and Gunnar Andersson
27. Clear and Turbid Water in Shallow Norwegian Lakes Related to
Submerged Vegetation
361
Bj0m A. Faafeng and Marit Mjelde
28. Macrophytes and Turbidity in Brackish Lakes with Special
emphasis on the Role of Top-Down Control
369
Erik Jeppesen, Martin S0ndergaard, Jens Peder Jensen, Eva Kanstrup,
and Birgitte Petersen
Contents
x
Section 3. Interdisciplinary Discussions
29.
30.
31.
Structuring Role of Macrophytes in Lakes: Changing Influence
Along Lake Size and Depth Gradients
Avital Gasith and Mark V. Hoyer
381
Nutrient-Loading Gradient in Shallow Lakes:
Report of the Group Discussion
Stephen R. Carpenter, Ellen van Donk, and Robert G. Wetzel
393
Alternative Stable States
Marten Scheffer and Erik Jeppesen
Index
397
407
Contributors
Gunnar Andersson
County Administration Board, S-205 15
Malmo, Sweden
John W. Barko
Environmental Laboratory, USACE
Waterways Experiment Station,
Vicksburg, MS 39180-6199,
USA
Irmgard Blindow
Department of Limnology, Lund
University, S-223 62 Lund, Sweden
Christer Bronmark
Department of Ecology, Lund University,
S-223 62 Lund, Sweden
Lise Bruun
Department of Lake and Estuarine
Ecology, National Environmental Research
Institute, DK-8600 Silkeborg, Denmark
Stephen R. Carpenter
Center for Limnology, University of
Wisconsin, Madison, WI 53706, USA
xi
xii
Contributors
Kirsten Christoffersen
Freshwater Biological Laboratory,
University of Copenhagen, DK-3400
Hiller0d, Denmark
Hugo Coops
Institute for Inland Water Management
and Waste Water Treatment, 8200 AA
Lelystad, The Netherlands
Greg Cronin
Cooperative Institute for Research in
Environmental Sciences (CIRES),
University of ColoradolNOAA, Boulder,
CO 80309-0216, USA
Larry B. Crowder
Marine Laboratory, Nicholas School of
the Environment, Duke University,
Beaufort, NC 28516-9721, USA
Paul Cunningham
Bureau of Fish Management, Department
of Natural Resources, Madison, WI
53703, USA
Sebastian Diehl
Zoologisches Institut,
Ludwig-Maxirnilians UniversiHit,
D-80333 Miinchen, Germany
Bj0m A. Faafeng
Norwegian Institute for Water Research,
Kjelsaas, 0411 Oslo, Norway
Adrienne 1. Froelich
Department of Biological Sciences,
University of Notre Dame, Notre Dame
IN 46556, USA
Sarig Gafny
Institute for Nature Conservation Research,
George S. Wise Faculty of Life Sciences,
Tel Aviv University, Tel Aviv 69978,
Israel
Avital Gasith
Institute for Nature Conservation Research,
George S. Wise Faculty of Life Sciences,
Tel Aviv University, Tel Aviv 69978, Israel
Binhe Gu
Division of Environmental Sciences,
St. 10hns River Water Management
District, P.O. Box 1429, Palatka, FL
32178-1429, USA
Contributors
xiii
Anders Hargeby
Department of Biology, University
College of Karlstad, S-65188 Karlstad,
Sweden
Mark V. Hoyer
Department of Fisheries and Aquatic
Sciences, University of Floridallnstitute
of Food and Agricultural Sciences,
7922 NW 71st Street, Gainesville, FL
32653-3071, USA
William F. James
Eau Galle Aquatic Ecology Laboratory,
USACE Waterways Experiment Station,
Spring Valley, WI 54767, USA
Jens Peder Jensen
Department of Lake and Estuarine
Ecology, National Environmental
Research Institute, DK-8600 Silkeborg,
Denmark
Erik Jeppesen
Department of Lake and Estuarine
Ecology, National Environmental
Research Institute, DK-8600 Silkeborg,
Denmark
John Iwan Jones
Royal Halloway Institute for
Environmental Research, Royal
Halloway College, University of London,
Huntersdale, Callow Hill, Virginia Water,
Surrey GU25 4LN, UK
Klaus JUrgens
Max Planck Institute for Limnology,
D-24302 PIOn, Germany
Timo Kairesalo
Department of Ecological and
Environmental Sciences, University of
Helsinki, 15210 Lahti, Finland
Eva Kanstrup
Department of Lake and Estuarine
Ecology, National Environmental Research
Institute, 8600 Silkeborg, Denmark
XIV
Contributors
Ryszard Komij6w
Department of Hydrology and
Ichthyobiology, University of
Agriculture, 20-950 Lublin 1, Poland
Torben L. Lauridsen
Department of Lake and Estuarine
Ecology, National Environmental Research
Institute, DK-8600 Silkeborg, Denmark
David M. Lodge
Department of Biological Sciences,
University of Notre Dame, Notre Dame,
IN 46556, USA
Thomas H. Martin
School of Forest Resources,
Pennsylvania State University, University
Park, PA 16802, USA
Elizabeth W. McCollum
1715 Broadview Lane, Apt. 209, Ann
Arbor, MI 48105, USA
Marie-Louise Meijer
Institute for Inland Water Management
and Waste Water Treatment, 8200 AA
Lelystad, The Netherlands
Stuart F. Mitchell
Department of Zoology, University of
Otago, Dunedin, New Zealand
Marit Mje1de
Norwegian Institute for Water Research,
Kjelsaas, 0411 Oslo, Norway
Robert E. Moeller
Department of Earth and Environmental
Science, 31 Williams Drive, Lehigh
University, Bethlehem, PA 18015, USA
Brian Moss
School of Biological Sciences,
University of Liverpool, Liverpool
L69 3BX, UK
Nathan Nibbelink
Department of Zoology and Physiology,
University of Wyoming, Laramie, WY
82071, USA
Contributors
xv
Mark Olson
Cornell University Biological Field
Station, Bridgeport, NY 13030, USA
Craig W. Osenberg
Department of Zoology, University of
Florida, Gainesville, FL 32611-8525,
USA
Tom Pellett
Bureau of Integrated Science Services,
Wisconsin Department of Natural
Resources, Madison, WI 53716, USA
Martin R. Perrow
ECON, Biological Sciences, University
of East Anglia, Norwich NR4 7TJ, UK
Lennart Persson
Department of Animal Ecology, Umea
University, S-901 87 Umea, Sweden
Birgitte Petersen
Department of Lake and Estuarine
Ecology, National Environmental
Research Institute, DK-8600 Silkeborg,
Denmark
Marten Scheffer
Wageningen Agricultural University,
Department of Water Quality
Management and Aquatic Ecology,
P.O. Box 8080, NL-6700 DD
Wageningen, The Netherlands
Claire L. Schelske
Department of Fisheries and Aquatic
Sciences, University of Floridallnstitute
of Food and Agricultural Sciences, 7922
NW 71 st Street, Gainesville, FL
32653-3071, USA
Louise Schliiter
The Water Quality Institute, Agem
Alle 11, DK-2970 H!3rsholm, Denmark
Jan Simons
Department of Ecology and
Ecotoxicology, Free University, 1081 HV
Amsterdam, The Netherlands
XVI
Contributors
Martin SliSndergaard
Department of Lake and Estuarine
Ecology, National Environmental
Research Institute, DK-8600 Silkeborg,
Denmark
Morten SliSndergaard
Freshwater Biological Laboratory,
University of Copenhagen, DK-3400
HillerliSd, Denmark
Christine Storlie
Bureau of Integrated Science Services,
Wisconsin Department of Natural
Resources, Madison, WI 53716, USA
Jon Theil-Nielsen
Freshwater Biology Laboratory,
University of Copenhagen, DK-3400
HillerliSd, Denmark
Anett Trebitz
U.S. Environmental Protection Agency,
Duluth, MN 55804, USA
Marcel S. Van den Berg
Department of Ecology and
Ecotoxicology, Free University, 1081 HV
Amsterdam, The Netherlands
Ellen Van Donk
Netherlands Institute of Ecology, Centre
for Limnology (NIOO-CL),
Rijksstraatweg 6, 3631 AC Nieuwersluis,
The Netherlands
Jan E. Vennaat
International Institute for Infrastructural,
Hydraulic and Environmental
Engineering, 2601 DA Delft,
The Netherlands
Robert T. Wass
Department of Zoology, University of
Otago, Dunedin, New Zealand
Robert G. Wetzel
Department of Biological Sciences,
University of Alabama, Tuscaloosa, AL
35487-0206, USA
Contributors
Karen Wilson
Center for Limnology, University of
Wisconsin, Madison, WI 53706, USA
Johnstone O. Young
School of Biological Sciences,
University of Liverpool, Liverpool
L69 3BX, UK
xvii
1.
2.
3.
4.
5.
6.
7.
8.
Stuart F. Mitchell
John Iwan Jones
Stephen R.Carpenter
Morten SjIlndergaard
Klaus JUrgens
Brian Moss
Robert G.Wetzel
Sebastian Diehl
9.
10.
11.
12.
13.
14.
15.
16.
Erik Jeppesen
Greg Cronin
Christer Bronmark
Torben L. Lauridsen
Mark V. Hoyer
Bjjllm Faafeng
Jens Peder Jensen
Ellen van Donk
17.
18.
19.
20.
21.
22.
23.
24.
Irmgard Blindow
Jan E. Vermaat
David M. Lodge
Marie-Louise Meijer
Lennart Persson
Kirsten Christoffersen
Lene Jacobsen
Larry B. Crowder
25.
26.
27.
28.
29.
30.
Timo Kairesalo
Martin SjIlndergaard
Ryszard Kornij6w
Marten Scheffer
Avital Gasith
John W. Barko
1. General Themes
1.
Fish-Habitat Interactions Mediated via
Ontogenetic Niche Shifts
Lennart Persson and Larry B. Crowder
Introduction
A fundamental characteristic of fish is that individuals increase in size by several
orders of magnitude over their ontogeny (Werner, 1988). This increase in size
generally means that the individual changes its food resource use during development. The change in resource use can take many different routes involving
changes between carnivory and herbivory/detritivory (Gerking, 1994). Commonly, an increase in prey size eaten is observed in connection with the increase in
consumer size, which potentially involves a change from zooplanktivory, to benthivory, and ultimately, to piscivory (Persson, 1988; Osenberg et al., 1994; Olson
et al., 1995). These changes in resource use are, in turn, generally associated with
habitat shifts in which complex habitats such as vegetated areas of lakes may
function both as a resource base and as a refuge from predation (Heck and
Crowder, 1991; Mittelbach and Osenberg, 1993; Persson, 1993; Olson et aI., 1995;
Persson and Ekl6v, 1995).
The focus of this chapter is on how size-dependent processes in fish interact
with habitat structure to shape ecological communities. Although we will restrict
our treatment to fish-habitat interactions, changes in body size over ontogeny and
consequences thereof for ecological interactions are by no means restricted to fish
but are rather ubiquitous in aquatic environments (Mittelbach et al., 1988; Persson,
1988; Stein et aI., 1988). For example, zooplankton and macroinvertebrate species
generally increase substantially in size over their lifetime, an increase that often
3
4
L. Persson and L.B. Crowder
involves metamorphosis (Werner, 1988). Correspondingly, size-dependent interactions have been shown to have substantial effects on population dynamics in
Daphnia (McCauley and Murdoch, 1990), and major effects of size-structured
interactions for overall community dynamics have, for example, been demonstrated in the Chaoborus larvae (Neill, 1988).
The impact of vegetation on fish is multifold. Vegetation offers a physical
structure that affects both competitive and predatory interactions between different species and sizes of fish (Winfield, 1986; Diehl, 1988; Persson, 1991).
Vegetation is also associated with high densities of invertebrate prey, which have
been shown to affect food consumption and growth of the fish (Crowder and
Cooper, 1982; Heck and Crowder, 1991; Diehl, 1993; Persson, 1993; Diehl and
Kornijow, this volume, Chapter 2). Macrophytes and associated epiphytic algae
may also form a resource for fish (Prejs, 1984; Hansson et aI., 1987). Fish have
feedback effects on vegetation by their direct consumption of vegetation and
indirectly via other trophic components or abiotic routes such as sedimentfeeding-induced turbidity. Fish may therefore also affect habitat structure. Several
of the indirect effects of fish on vegetation are considered in other chapters of this
volume. We therefore largely restrict our treatment to how habitat structure and
associated resources affect fish performance, although we include a discussion on
the importance of fish for nutrient fluxes between habitats.
Because size plays such a prominent role in interactions among fish species and
for fish-habitat interactions, we first consider basic ecological capacities of individual fish in relation to size and discuss how these size-dependent capacities are
influenced by habitat structure. The ecological performances of individuals in
different habitats are also affected by species-specific characteristics, and we
cover some of these characteristics. Based on the individual level characterizations, we review how life history phenomena related to ontogenetic constraints
affect the performance of different functional groups of fish, specifically piscivore-prey fish interactions. We also discuss the implications of ontogenetic
changes for overall community, ecosystem, and nutrient dynamics. Finally, we
point to how our understanding of interactions between fish and habitat structure
may be enhanced by the application of stage-structured population models.
Size-Dependent Foraging and Predator Avoidance Abilities
and Habitat Structure
Increasing in size imposes a series of constraints on the organism, ranging from
physical and mechanical constraints to ecological constraints (Miller et aI., 1988;
Werner, 1988). An example of the former is that the Reynolds number that a fish
larva experiences is quite different from that an adult large fish experiences
(Webb, 1978). Examples of the latter are that the foraging and predator avoidance
capacities are closely related to the size of the fish. With respect to foraging rate,
an increase in size means that two basic components of the individual's competitive ability (i.e., its foraging rate and its metabolic demands) change (Peters, 1983;
1. Fish-Habitat Interactions
A
5
c
B
~ ~---------------,
t:
~
/:~~eg/:~
...
\
as
E
.••......................•..................
1E:::I
....... Pelagic'i::
--Vegetation
....... Pelagic
~ -\-,-.."""......,~:;:;;:;;;;=~~
Body mass
Body mass
Body mass
Figure 1.1. (A) General relationship between foraging gains and metabolic costs and body
mass. (B) Attack rate as a function of body mass in the benthic, vegetation, and open water
habitats based on encounter rates for bluegill sunfish. (Data from Mittelbach, 1981.) The
attack rate has been assumed to have a hump-shaped relationship with body mass (see
Persson et aI., submitted). For the clarity of presentation, the maximum attack rate has been
assumed to be the same for all prey types. (C) The minimum resource requirements
necessary for maintenance as a function of body mass in three habitats based on the
encounter rate function in B. For the clarity of presentation, it has been assumed that the
prey weights in the different habitats are the same. Observe the log axes in B and C.
Calder, 1984; Miller et al., 1988; Werner, 1988). Metabolic demands as a function
of body weight are generally assumed to be described by a power function with a
slope varying between 0.6 and 0.9 (Peters, 1983; Calder, 1984; Werner, 1988)
(Fig. 1.1 A). The capacity to ingest energy has also been described by a power
function of body size. However, for a prey of a specific size, the foraging rate is
not expected to increase monotonically with size but to increase to a maximum to
thereafter decrease (Tripet and Perrin, 1994; Persson et al., submitted) (Fig. 1.1B).
The form of this general function has been substantiated in fish larvae (Bailey and
Houde, 1989), and a review of size-dependent interactions in freshwater and
marine fish larvae is given in Miller et al. (1988). The decreasing part of the
relationship relates, among other things, to the capacity of an individual to discern
small prey and make fine-tuned maneuvers, which decreases with body size
(Breck and Gitter, 1983; Noakes and Godin, 1988; see also Persson et al.,
submitted).
The rate by which the foraging capacity increases with body size, generally
termed the ontogenetic scaling of foraging rate, varies among taxa (Wilson, 1975;
Werner, 1988, 1994; Lundberg and Persson, 1993; Persson et al., submitted). This
variation is partly related to differences in foraging methods used by different
functional groups of consumers. For example, the ingestion rate of filter feeders is
expected to scale to body size with a higher slope than that of particulate feeders.
The ontogenetic scaling of foraging rate to body size will also vary within taxa
based on habitat-specific constraints on search behavior. The slope of the size
6
L. Persson and L.B. Crowder
Body mass
Figure 1.2. General relationship between mortality rate and body mass in the open water (solid
line) and vegetation (dashed line) habitats. Due
to the presence of predation from other small
(albeit larger) fishes that are confined to the
vegetation due to their own predation risk and
potentially invertebrate predators, the form of
the mortality rate function in the vegetation can
change so that mortality is highest for the smallest stages (dotted + dashed lines).
scaling of foraging rate is expected to be higher for a fish foraging in a threedimensional environment such as the pelagic habitat than for a fish foraging in a
two-dimensional environment such as the benthic habitat. Correspondingly, it has
been found that the slope of the relationship between the attack rate and body
weight of the bluegill sunfish (Lepomis maeroehirus) decreases from pelagic prey
to vegetation prey to benthic prey (Mittelbach, 1981) (Fig. l.1B). Mechanistic
explanations for why small and large fish should be differently affected by vegetation are that the encounter rate and swimming speed advantages of larger fish will
decrease in structured habitats. The size scaling of foraging rate is thus expected to
decrease for fish moving from a pelagic habitat to a vegetation habitat, which, in
tum, will affect the competitive abilities of differently sized individuals (Persson
et aI., submitted) (Fig. 1.1 C). This suggested effect of vegetation structure on the
size scaling of foraging intake is supported by other studies. Ryer (1987, 1988), for
example, showed that the amount of prey encountered and consumed by large
pipefish (Syngnathus fuse us ) decreased in eelgrass (Zostera marina), whereas the
foraging efficiency of small individuals was unaffected by the amount of structure.
Several studies have shown that the growth rate of prey fish is retarded when they
are confined to the vegetation due to predation risk (Mittelbach, 1988; Persson,
1993; Diehl and Eklov, 1995; Persson and EklOv, 1995). This suggests that, in
addition to a decrease in the slope of the size scaling of foraging rate to body size,
the maximum foraging rate decreases when the fish shifts from the pelagic to the
vegetation habitat (note that the maximum foraging rate is assumed to be habitat
independent in Fig. l.1B and C).
The risk of being consumed by predators is also strongly connected to body
size. Until the late 1980s, it was generally thought that the predation mortality rate
of larval fish decreased monotonically with increasing prey fish size (Gilliam,
1982; Fuiman and Magurran, 1994) (Fig. 1.2). However, this expectation was
based on studies that focused on the capture success (ratio of prey consumed to
prey attacked) of predators and neglected other parts of the predation cycle such as
the encounter rate (Fuiman, 1994; Fuiman and Magurran, 1994). Once these parts
of the predation cycle were considered, it has been predicted and confirmed that
the vulnerability of larval fish to raptorial predators increases to a maximum and
then decreases as prey fish size increases (Bailey and Houde, 1989; Fuiman, 1989;
Litvak and Leggett, 1992; Pepin et aI., 1992) (Fig. 1.2). This pattern has been
1.
Fish-Habitat Interactions
7
suggested to result from an increase in the encounter rate between predator and
prey due to increased swimming speeds and increased pigmentation of the fish
prey and a simultaneous decrease in capture success of predators due to better
escape responsiveness of the prey fish as they grow and develop (Fuiman and
Magurran,1994).
Complex habitats like vegetated habitats may affect the relationship between
body size and predation mortality of prey fish by lowering the overall predation
efficiency ofpiscivores (Savino and Stein, 1982; Heck and Crowder, 1991; Christensen and Persson, 1993; Persson and Ekl6v, 1995) (Fig. 1.2). The mechanisms
behind the decreased predation efficiency in complex habitats can be both
decreased encounter rate between predator and prey and decreased capture success of the predator once the prey has been encountered (Andersson, 1984; Main,
1987; Savino and Stein, 1989a,b; Christensen and Persson, 1993). Complex habitats may also affect the form of the relationship between body size and predation
mortality (Fig. 1.2). For example, the size-dependent mortality in the bluegill
sunfish can take the form of a monotonically decreasing function rather than a
hump-shaped one. The high predation mortality of very small stages of bluegill in
the vegetated habitat is a consequence of predation from other small (albeit larger)
fishes, which are confined to the vegetation due to their own predation risk, and
potential invertebrate predators that prey on the very youngest stages (Werner and
Hall, 1988; G. Mittelbach, personal communication) (Fig. 1.2). The monotonically
decreasing form of the predation mortality function in the vegetation habitat may
be one reason (in addition to differences in resource size distributions between
vegation and pelagic habitats) why the smallest stages of bluegills spend the first
few weeks after hatching in the pelagic habitat before returning to the vegetation
habitat (Werner and Hall, 1988; G. Mittelbach, personal communication).
Predator-Induced Habitat Shifts: Competitive and Predator
Avoidance Abilities of Different Fish Species
The presence of piscivores affects the habitat use of small stages of fish and often
restricts them to the littoral vegetated habitat (Mittelbach, 1986, 1988; Turner and
Mittelbach, 1990; Persson, 1991, 1993; Tonn et aI., 1992; Brabrand and Faafeng,
1993; Persson and Ekl6v, 1995). This will generally lead to an increased competition intensity among refuging prey fish (Mittelbach, 1988; Turner and Mittelbach,
1990). Predator-mediated habitat use may also release invulnerable size classes
from competition from smaller vulnerable size classes (predator-mediated habitat
segregation) (Werner et al., 1983; Gilliam and Fraser, 1988; Savino and Stein,
1989a,b; Tonn et al., 1992; Christensen and Persson, 1993; Diehl and Ekl6v, 1995;
Carpenter et aI., this volume, Chapter 11).
How seeking refuge in vegetation habitats affects the competitive abilities of
prey fish depends on the species concerned. In their classic work on North
American centrarchids, Werner and Hall (1979) showed species-dependent foraging abilities of three sunfish species in the vegetation, which resulted in differen-
8
L. Persson and L.B. Crowder
ces in the timing of niche shifts between the three species. Studies of European
species have shown that different fish species are affected differently by the
presence of vegetation. Winfield (1986) found that the foraging rate of the cyprinid rudd (Seardinius erythrophthalmus) on Daphnia was only affected at stem
densities greater than 200/m2 and that the foraging rate of juvenile perch (Perea
fluviatilis) did not decrease even at the highest stem density used (600 stems/m2 ).
By contrast, the foraging rate of roach (Rutilus rutilus) decreased substantially
even at the lowest stem density used. Similar results were obtained by Diehl
(1988) in a study of the foraging efficiencies of roach, bream (Abramis brama),
and perch feeding on chironomids (i.e., the foraging performances of roach and
bream decreased strongly in the presence of vegetation whereas the foraging
performance of perch was only slightly affected by vegetation).
Because the effects of habitat structure on foraging performance are species(and size-) specific, predator-induced habitat shifts by prey fish affect and even
reverse the outcome of competitive interactions among refuging prey fish. Persson
(1991) showed that juvenile perch and roach responded to the presence of piscivorous perch by moving into the vegetation refuge. This resulted in a reversal of
the foraging advantage of roach (in the open water habitat) to a foraging advantage
of perch (in the vegetation refuge). The shift in relative foraging performance, in
tum, resulted in changes in relative growth rates. Although the growth rates of
roach were higher than those of perch in the absence of piscivores, this relationship was reversed in the presence of piscivorous perch. The interactions between
refuging juvenile roach and perch are thus affected by the structure per se. Because
vegetation structure is also associated with vegetation-attached resources (see
Diehl and Kornij6w, this volume, Chapter 2), the prey communities inhabiting
vegetation may have additional effects on the competitive interactions among fish
species. Perch have been shown to be superior foragers to roach on macroinvertebrates (Persson, 1988). Correspondingly, Persson (1993; see also Persson and
EklOv, 1995) found that invertebrate resources associated with vegetation structure additionally competitively favored juvenile perch over roach in vegetation
refuges as reflected in both diet and growth patterns. Because perch and roach
make up most of the total fish biomass in many European lakes, the effects of
vegetation structure on the interactions between these two species will have
ramifications for overall community and lake ecosystem dynamics. Roach as a
competitor with juvenile perch may severely limit the recruitment of perch to large
piscivorous stages in the absence of vegetated habitats (see below).
Different species of prey fish do not only vary in their habitat specific foraging
capacities but also in their habitat-specific abilities to avoid predation. Christensen
and Persson (1993) found that juvenile roach were more efficient in avoiding
piscivorous perch than juvenile perch in both open water and simulated vegetation
consisting of strings (see also Persson and Ekl6v, 1995). However, juvenile perch
were more efficient in avoiding predators by using crevices. When simultaneously
offered perch and rudd in field enclosures, pike captured more rudd than perch in
environments lacking vegetation, whereas the opposite was the case in environments with vegetation (Ekl6v and Hamrin, 1989). Interactions between pisci-
l. Fish-Habitat Interactions
9
vorous predators and prey fish are also affected by the type of predator species
present. Pike (Esox lucius) have been found to be a more efficient predator than
perch and pikeperch (Stizostedion lucioperca) in vegetation, whereas pikeperch
and perch are more efficient in open water (Ekl6v, 1992; Ekl6v and Diehl, 1994;
Greenberg et aI., 1995). The presence of vegetation will also affect the foraging
mode of specific piscivorous predators. For example, perch and largemouth bass
(Micropterus salmoides) change from an active pursuit foraging mode to an
ambush sit-and-wait foraging mode with an increase in vegetation density (Savino
and Stein, 1982; EklOv and Diehl, 1994).
Habitat Shifts and Mixed Competition-Predation Interactions and
Ontogenetic Constraints
In the previous sections, we pointed out the importance of size when studying
interactions among fish populations and that habitat structure affects the size
scaling of the performance of the fish in terms of foraging capacity and predator
avoidance ability. In this section, we consider how growth in size imposes changes
in the nature of ecological interactions and also puts constraints on fish life history.
The latter relates to the fact that the most efficient morphology for handling prey
varies with prey types used over ontogeny.
As a result of variability in fish individual growth rates and the presence of
several size cohorts in fish populations, interactions among fish species are characterized by a mixture of competitive and predatory interactions (Werner et aI.,
1983; Mittelbach, 1986, 1988; Werner, 1986; Persson, 1988; Persson and Greenberg, 1990). This mixture of competitive and predatory interactions takes place at
several temporal and spatial scales. On a short time scale and within a system
spatial scale, the behavioral decisions of an individual fish as a function of its size
are a result of both competitive and predatory considerations (Gilliam, 1982; Lima
and Dill, 1990). Behavioral models have predicted that, given different foraging
returns and predation risks in different habitats (i.e., open water versus vegetated
habitats), juvenile fish are expected to choose the habitat with the lowest mortality
rate/growth rate ratio (assuming equilibrium and no time constraints) (Gilliam,
1982). This prediction (or predictions analogous to this) has been supported in
several experimental studies (Gilliam and Fraser, 1988; Turner and Mittelbach,
1990). On a longer time scale, interactions among species may change between
competitive interactions and predatory interactions as a result of individual growth
(Wilbur, 1988). In fish, these changes between mainly competitive interactions to
mainly predatory interactions are often associated with changes in habitat use in
which the vegetation habitat plays a crucial role. For example, although juvenile
fish of different species often compete for resources when refuging from predators
in vegetation habitats, one of the species may start to prey on the other as they
increase in size and shift habitat (Mittelbach, 1986; Werner, 1986; Olson et al.,
1995; Persson and Ekl6v, 1995). The importance of predatory versus competitive
interactions may also vary among systems mediated through the size structures of
10
L. Persson and L.B. Crowder
the fish populations (Persson, 1988; Persson and Greenberg, 1990). In this case,
the availability of submerged vegetation has been advanced as a major factor
influencing the role of competitive and predatory interactions (Persson, 1988; see
below).
The presence of size-dependent ontogenetic niche shifts imposes a series of
constraints on the organism as a result of ontogenetic covariance (Werner, 1988).
Natural selection will operate on morphological and behavioral traits over the
whole life cycle of the individual fish and traits that are optimal at one ontogenetic
niche are suboptimal in other ontogenetic stages (Werner and Hall, 1979; Werner,
1986, 1988; Persson, 1988). An illustrative example is an adult piscivorous species
that as a small planktivore will be burdened with morphologies and behaviors
more adapted for piscivory than for planktivory. The morphological traits for a
typical particulate feeding planktivore are, for example, a compressed body and a
small gape size, which will allow the fish to capture relatively small and nonevasive prey items efficiently at high swimming speed (Werner, 1977; Webb,
1984). By contrast, piscivorous feeding involves large evasive prey, which requires traits such as high attack speed, large gape size, and attacking foraging
mode (Webb, 1984). Based on the constraints imposed by ontogenetic covariance,
a general hypothesis has been advanced. This hypothesis states that a species
undergoing substantial ontogenetic niche shifts during its life will be a less
efficient predator on small zooplankton prey compared with a species undergoing
less drastic niche shifts such as a planktivore specialist (Werner, 1986; Persson,
1988). Experimental support for this hypothesis has been provided for at least two
species constellations-the perch-roach interaction and the largemouth bassbluegill interaction (Werner, 1977; Persson, 1988). For perch, several aspects of
the body morphology and behavior are generally associated with benthivorous
feeding in vegetated habitats rather than planktivory or piscivory. These aspects
include a relatively low cruising speed, a relatively deep body, laterally inserted
pectoral fins, and enlarged dorsal fins (Ekl6v and Persson, 1995). We expect that
the support for the hypothesis of an ontogenetic trade-off cost in piscivorous
species will increase when experimental data for additional species constellations
are provided.
As a consequence of ontogenetic trade-offs in piscivores, Persson (1988) suggested that the interactions between piscivores and planktivores are characterized
by a high degree of asymmetry. By definition, piscivores have a predatory advantage, which is counteracted by a competitive advantage on juvenile resources
for planktivores due to ontogenetic trade-off costs in the piscivore. This type of
asymmetric interaction, and particularly changes in the relative strength of the
predatory versus competitive advantage has been suggested to have major consequences for overall community and ecosystem dynamics (Persson et aI., 1991,
1992). In moderately productive systems, the proportion of piscivores (piscivorous perch making up most of piscivore biomass) of total fish biomass is high,
composing up to 80% of total fish biomass (Fig. 1.3). By contrast, the proportion
of piscivores of total fish biomass in highly productive systems is low (:$;20%) as
a result of a severe bottleneck in the recruitment of juvenile perch to piscivorus
11
1. Fish-Habitat Interactions
Figure 1.3. Changes in percentage
pelagic piscivorous perch biomass of
total pelagic fish biomass and phytoplankton biomass (chlorophyll a in
~g/L) along a phosphorus-loading
gradient in Swedish lakes. (Data
from Persson et al., 1991; Carpenter
et al., 1996.)
80.-------~~--------------,
<..>
~~
~o.
.......
60
11)
'5.15
L;"-
40
0.0
e·~
II)
.Q
L; . -
20
00.
O+-~~~~--~.~~.---~~~
0.01
0.1
1
10
P loading (g/m2-year)
stages caused by planktivorous and benthivorous cyprinids (Persson, 1988; Persson and Greenberg, 1990). These changes in the importance of piscivores also
have feedback effects on other trophic levels including zooplankton and phytoplankton. For example, a tenfold increase in phosphorus loading from 0.03 to
0.3 g/m2 a year only led to minor increase in phytoplankton biomass (Carpenter et
aI., 1996), which suggests that an increased piscivore biomass may prevent an
increased phosphorus loading from being expressed as an increase in phytoplankton biomass within this range of phosphorus loadings (Fig. 1.3). By contrast,
in highly productive systems with a low proportion of piscivores, phytoplankton
biomass increases steadily with phosphorus loading.
Studies of fish communities show that a shift in the species numerically
dominating the fish community takes place along the productivity gradient. This
shift involves a change in dominance of percids (mainly perch) in mediumproductive lakes to a dominance of cyprinids in highly productive lakes, which is
also reflected in changes in size structures of the populations (Persson, 1988).
Although correlated to productivity, this major change in fish community structure, which involves feedback effects on lower trophic levels, has been hypothesized to be also affected by changes in the availability of submerged vegetation
with increasing productivity (Persson et aI., 1992; Persson, 1993; Persson and
Eklov, 1995). This hypothesis is related to the observation that the importance of
submerged vegetation generally is at a maximum in moderately productive lakes
(Wetzel, 1979), where also piscivore biomass has a maximum. Mechanistic explanations for why vegetation structure should affect piscivore-planktivorelbenthivore fish
interactions are that the performance of the juveniles of the major piscivore, perch, in
relation to competing planktivores is strongly related to the presence of vegetation
structure (see above; Diehl and Kornij6w, this volume, Chapter 2).
Habitat Structure and Stage-Structured Interactions
in Lakes: Two Examples
In many lakes throughout the north central United States, the Centrarchid bluegill
sunfish make up most of total fish biomass (Osenberg et al. 1988, 1994; Mittel-
12
L. Persson and L.B. Crowder
bach and Osenberg, 1993). This species hatches in the littoral and moves to open
water for a few weeks before moving back to the sheltered vegetation habitat. As
an adult, it feeds on zooplankton in open water, and the body size at the shift to this
habitat depends on predation risk from largemouth bass (Mittelbach and Chesson,
1987; Werner and Hall, 1988). Mittelbach and Osenberg (1993; see also Osenberg
et al., 1994) have suggested that the limnetic productivity of zooplankton sets the
limit to the production (including fecundity) of adult bluegill sunfish, which, in
tum, determines the intensity of competition in the littoral vegetation habitat
through juvenile bluegill recruitment. A negative effect of juvenile bluegill sunfish
density on the growth of its competitors, including the major adult piscivorous
predator largemouth bass, has been demonstrated in cross-lake comparisons as
well as in pond/enclosure experiments (Mittelbach, 1988; Osenberg et aI., 1994;
Olson et aI., 1995). The strong competitive effect of juvenile bluegills on other
refuging littoral fish can partly be related to the fact that juvenile bluegill outnumber the other species. For the interaction between largemouth bass and bluegill, it
has also been experimentally demonstrated that the per capita effect of juvenile
bluegill on young of the year (YOY) bass is larger than the reverse (Olson et aI.,
1995). This was the case despite a substantial resource partitioning between
juvenile bluegill and bass, because bluegill caused changes in the size structure of
major invertebrate prey (bluegill fed on smaller shared prey than bass and prevented these resources from growing to the larger sizes used by bass).
The interaction between largemouth bass and bluegill is a typical example of a
mixture of competitive and predatory interactions that also involves habitat shifts.
The effect of bluegill density (both adult and juvenile) on YOY largemouth bass
growth has been found to be negative (competitive interactions), whereas the
effect of YOY bluegill density on the growth of large largemouth bass is positive
(predator-prey interaction) (Olson et aI., 1995). The effects of bluegill density on
adult bass density is also positive, and as a higher growth rate of adult bass leads
to higher per capita fecundity, also YOY largemouth bass density is positively
related to juvenile bluegill density (Fig. 1.4) (see also below).
In many Scandinavian lakes, perch and roach are the two dominating species.
Roach are efficient zooplanktivores competing with juvenile perch but may also
feed on macro invertebrates and non animal food items. Perch are ontogenetic
omnivores and start to feed on zooplankton, to thereafter shift to macroinvertebrates to finally become piscivorous (Persson, 1988). Vegetation has, as was
considered above, been shown to affect both competitive interactions between
roach and perch and predator-prey interactions between piscivorous perch and
small roach and perch. The macroinvertebrate feeding stage has been identified to
be an important bottleneck in the recruitment of perch to piscivorous stages
(Persson, 1986, 1988), and increased availability of habitats with submerged
vegetation is likely to decrease the limitations set by this recruitment bottleneck
(Diehl and Kornij6w, this volume, Chapter 2).
In contrast to the largemouth bass-bluegill density relationships, the density
relationship between perch and roach can be both positive and negative, which has
been suggested to be related to the availability of submerged vegetation (see
13
l. Fish-Habitat Interactions
Largemouth bass-Bluegill
ill
:::>
c..
~
8.------------------.
•
6
•
~
'iii
lij 4
b
~
~
'iii
c
en
gj
•
QJ
e'
> O+----,----r----r--~
10
20
30
40
o
ttl
....J
•
•
0.4
•
•
•
0.8
.0
2
•
1.2
"C
•
en
gj
w
c..
:::>
Q)
"C
.0
•
•
~
O+---~----'----r--~
o
Bluegill density (CPUE)
10
20
30
Bluegill density (CPU E)
40
Perch-Roach
-
ill
1t 0.8
•
~
~ 0.6
•
~
•
:0
.s::
0.4
•
•
QJ
c..
O·
o
•
0.2
•
~
en
en
8
6
•
ttl
E 4
0
:0
.s::
!:
QJ
••
!: 0.2
ill
:::>
c..
2
•••
c..
0.4
0.6
0.8
Roach biomass (CPU E)
•
•
•
•
O+---'---.---·.---r-~
o
0.5
1
1.5
2
2.5
Roach biomass (CPUE)
Figure 1.4. (Top) Density catch per unit effect [CPUEj relationships between YOY largemouth bass and bluegill (left) and between large largemouth bass and bluegill (right) in
Michigan lakes. (Data from Olson et aI., 1995). (Bottom) Biomass (kg CPUE) relationships
between roach and perch in low to medium productive lakes (left) and medium to highly
productive systems (right). (Data from Persson et aI., 1991.)
above) (Fig. 1.4). This difference in density relationship between perch and roach
versus bass and bluegill can be related to differences in the life histories of the two
piscivores (Olson et al., 1995). Bass eventually become piscivorous in their fIrst year,
whereas it may take several years for perch to reach piscivorous stages. The potential
for competing prey to affect perch recruitment to piscivorous stages is thus higher.
Littoral-Pelagic Coupling in Lakes: Effects of Fish on
Nutrient Fluxes
It has long been recognized that littoral habitats are linked biogeochernically to the
open waters of lakes (Wetzel, 1979; Barko and James, this volume, Chapter 10).
14
L. Persson and L.B. Crowder
Table 1.1. Summary of Potential Mechanisms by Which Fish May Affect Macrophyte
Abundance, Littoral-Pelagic Couplings, and Overall Lake Dynamics
I. Routes for effects on submerged vegetation (habitat structure) by fish feeding
activities in macrophyte habitats
• Feeding on macrophytes
• Feeding on epiphytes
• Feeding on macroinvertebrates
• Feeding-induced uprooting of plants
II. Routes for effects on submerged vegetation (habitat structure) by feeding activities in
open water
• Zooplankton predation induced changes in phytoplankton biomass (transparency)
• Sediment feeding induced fluxes of nutrients to open water affecting
phytoplankton biomass (transparency)
III. Other fish induced coupling of littoral and pelagic habitats
• Transport of nutrients/organic matter to pelagic due to littoral feeding and pelagic
excretionlegestion
• Transport of nutrients/organic matter to littoral due to pelagic feeding at night by
in daytime refuging juvenile fish
• Recruitment of juvenile fish to the vegetation habitat from pelagic larval stages
over ontogeny
• Recruitment of adult fish to the pelagic habitat from the vegetation habitat over
ontogeny
Littoral zones are extremely productive; in addition to macrophyte production,
epiphyte production is now known to be substantial (Wetzel, 1990; Wetzel and
S0ndergaard, this volume, Chapter 7). Biomass, production, and diversity of
littoral invertebrates often exceed that of invertebrates in open water areas. Unfortunately, most studies of food web interactions in lakes have focused on either the
littoral or the pelagic habitat (Lodge et al., 1988). Linking littoral zone into the
whole lake requires us to consider its role as a refuge, habitat, and nutrient source
or sink. The role of food web links in the transformation or translocation of
nutrients between the littoral and pelagic habitats is not well known (Lodge et al.,
1988). We do know that fish often shift from the littoral to the pelagic habitat as
they increase in body size through ontogeny. Life history ornnivory in itself thus
means that there will be a coupling between different habitats (Table 1.1). For
example, in bluegill sunfish ontogenetic habitat shifts will couple the dynamics of
the invertebrate prey communities in the littoral zone with the zooplankton prey
community in the open water habitat (Osenberg et al., 1994). Furthermore, fish
move to and from the littoral habitat on a diel basis (Hall et al., 1979; Naud and
Magnan, 1988). Do these movements have implications for nutrients in pelagic
food webs?
Consumers may playa large role in lake nutrient budgets (Table 1.1). Although
limnologists initially focused on external loading of nutrients, for many systems
internal nutrient recycling contributes substantially to the nutrient budget. Nutrient
recycling and transformation by zooplankton are now widely recognized as impor-
1. Fish-Habitat Interactions
15
tant sources of nutrients for phytoplankton production (Vanni, 1996), but fish
effects on nutrient translocation and transformation and subsequent effects on
algal productivity and species composition have been overlooked until recently.
Only a few experimental studies document direct nutrient recycling by fish affecting phytoplankton community structure (Reinertsen et aI., 1986; Vanni and Findlay, 1990; Schindler, 1992; Vanni and Layne, 1997). In an experimental study,
Vanni et al. (1997) separate the community-level effects of nutrient recycling by
fish from those due to zooplankton and document that in lakes dominated by
planktivores, fish effects can exceed nutrient recycling effects of zooplankton on
algal community composition. Fish effects may be even more significant if one
considers that a substantial portion of fish diets is consumed in the littoral habitat
(Schindler et aI., 1993). In northern Wisconsin lakes, planktonic prey accounted
for less that 30% of diet biomass in planktivorous fishes; they mostly ate benthic
insects and periphyton. Even piscivorous fish consumed primarily benthic (65%)
or terrestrial (15%) prey; fish only accounted for 16% of the diet biomass of
piscivores (He and Kitchell, 1990; He and Wright, 1992; see also Brabrand et al.,
1990). But fish may move offshore and release nutrients (Schindler et aI., 1996;
Vanni, 1996). Excretion by consumers of nutrients derived from the littoral provides new nutrients for pelagic producers (Table 1.1).
Recent biogeochemical analyses of phosphorus cycles in the pelagic zones of
lakes suggest that observed levels of primary productivity are too high to be
supported by pelagic recycling alone. Caraco et al. (1992) estimated that more
than one-third of the pelagic primary production of Mirror Lake (NH) must be
supported by "new" phosphorus. Could the source of these nutrients be translocated by nutrients from the littoral? In recent reviews, both Vanni (1996) and
Schindler et al. (1996) argued strongly that pelagic nutrient budgets may only be
balanced by considering biologically driven phosphorus from the littoral. Our
current understanding suggests that the littoral zone is very likely a source of
nutrients for pelagic food webs in lakes. Both biogeochemical processes and
animal movements suggest that the pelagic habitat is basically a sink for littoral
productivity.
The relative importance of biological processes in nutrient translocation will
vary among lakes. Translocation of nutrients from the littoral or deep benthic
habitat (in stratified lakes) to the pelagic will vary with fish species composition as
well as the ontogenetic stage of the fish species. Species that feed exclusively in
the pelagic will contribute little to linkages from the littoral. Omnivorous fishes,
especially detritivores, will transport nutrients from the sediment and detritus to
the pelagic (Brabrand et aI., 1990; Mather et aI., 1995; Vanni, 1996). Benthic
feeding fishes transport nutrients from the bottom substrate back into the water
column, fueling pelagic production. Fish that migrate onshore-offshore (or among
habitat patches in shallow lakes) may increase nutrient translocation above that
accounted for by biogeochemical processes alone (Schindler et al., 1996). Lake
geomorphology will also contribute. Large lakes or those with limited shoreline
development and relatively small littoral zones (Gasith, 1991) may not foster
substantial littoral production or transfer from the littoral to the pelagic. Still,
16
L. Persson and L.B. Crowder
although Carpenter and Kitchell (1993) selected lakes for their experimental
manipulations of pelagic food webs that had minimal littoral habitat, the phosphorus budgets suggested strong inputs of phosphorus to the pelagic food webs
originating from the littoral habitats (Schindler et al., 1996).
Size-Structured Interactions, Habitat, and Population Dynamics
Given that fish movements on diel, seasonal, and ontogenetic scales can play a
large role in the translocation of nutrients, what methods do we have to predict
these behaviors? Early efforts used optimal foraging theory to forecast habitat
choice among individual juvenile fish (Mittelbach, 1981). Foraging strategies of
individuals, of course, also depend on the behaviors of other foragers. Game
theoretical approaches led to the prediction of an ideal free distribution among
competitors (Fretwell and Lucas, 1970; Milinski, 1979). Distributions of foragers
are also modified by the presence of predators (Werner et aI., 1983; Werner and
Gilliam, 1984; Abrahams and Dill, 1989), and new models of individual foraging
behaviors include the simultaneous optimization of several different objectives.
These models are fitness-based, and one of the most recent ones assumes that prey
fish choose habitats to maximize the ratio of net energy intake to probability of
death per unit time (Gilliam and Fraser, 1988). Werner and Hall (1988) used such
a model to predict the ontogenetic shifts in bluegills from the littoral to the pelagic
habitat as a function of predation risk in different lakes. These models assume that
risk of predation is static (i.e., the predators do not move). Recently, Hugie and
Dill (1994; see also Sih, in press) expanded the game theoretical approach to
predict habitat choice of both prey and predators in a two-habitat model with a
sedentary resource of the prey. One very interesting outcome of their model is that,
without behavioral interference among predators (e.g., when one would expect
predators to distribute themselves according to an ideal free distribution), the
density of prey in a habitat is determined only by the inherent riskiness of the
habitat and at the behavioral equilibrium is not influenced by habitat productivity
(i.e., productivity of the sedentary resource) per se. In other words, prey density is
only influenced by habitat features that influence risk of predation such as habitat
structure, turbidity, or light levels. In contrast to the prey, the habitat use of the
predator is affected by both habitat riskiness of the prey and productivity (see also
Diehl and Kornij6w, this volume, Chapter 2). When predators do interfere with
each other in the model, habitat productivity also plays a role in the expected
distribution of prey. The model of Hugie and Dill does not address dynamics on
the diel time scale, although one might expect die I behaviors to respond similarly
as light levels influence the risk of predation. Because availability of food
resources and riskiness of the habitat scale with habitat structure, submerged
macrophytes should playa large role in determining both the distribution of prey
and predators in littoral habitats.
Ideal free distribution models predict predator and prey habitat distribution on
a limited time scale and do not address long-term population dynamics, although
1. Fish-Habitat Interactions
17
the model of Hugie and Dill can be expanded to an equilibrium situation with no
net migration between habitats (see Oksanen et al., 1995). To handle population
dynamics, we need to derive models that include all relevant vital rates (growth,
birth, mortality, migration between habitats). To do this for size-structured populations such as fish is not an easy task. The least complex way to introduce size
structure is to use age-based demography. Mittelbach and Chesson (1987) argued
that for the Centrarchidae system, the dynamics of bluegill populations and interactions between bluegill and pumpkinseed (Lepomis gibbosus) sunfish may be
adequately characterized by an age-based model using two stages Uuveniles and
adults). It has been argued that an approach based on life history stages is sufficient for the description of populations if vital rates are similar within stages but
different between stages (Osenberg et aI., 1994).
The two life-stage interaction in bluegill is set up by piscivorous largemouth
bass in the open water that force the vulnerable size classes of bluegill to stay in
the littoral, vegetated area. The size at which bluegill move out to the open water
area varies among lakes depending on predation risk (i.e., largemouth bass density) (Werner and Hall, 1988). This flexible and stage-structured behavior of
bluegill leads to complex indirect effects between populations at several different
trophic levels that do not share the same habitat. In the two life-stage model
developed by Mittelbach and Chesson (1987), an increase in open water productivity is expected to lead to an increase in per capita adult fecundity. This will, in
tum, lead to an increased number of juveniles, which depletes the resource in the
vegetation refuge. As a result, per capita juvenile survival will decrease. The total
juvenile survival will still increase, causing an increase in adult numbers. Although adult density thus will increase with increased adult resource productivity,
density-dependent juvenile survival will prevent adult density from fully responding to the increase in their resource productivity. As a result of these
stage-structured interactions mediated via habitat shifts, there will be a positive
relationship between the densities of adults and adult resources and a negative
relationship between the densities of juveniles and juvenile resources. These
patterns contrast to predictions of standard predator-prey theory. The predictions
regarding how different stages of bluegill respond to an increase in productivity of
the adult resource as advanced by Mittelbach and Chesson (1987) have subsequently been supported by comparative field studies (Mittelbach and Osenberg,
1993).
Mittelbach and Chesson (1987) extended their stage-based model to include a
competitor of bluegill, the pumpkinseed sunfish. They showed that although adult
bluegill and adult pumpkinseed sunfish did not share resources (adult bluegill feed
on zooplankton and adult pumpkinseed feed on gastropods), an indirect negative
effect was present between these stages, mediated via interactions between
juveniles of both species in the vegetation refuge.
In many organisms, the ecological capacities in terms of foraging efficiency,
metabolic demands, capacity to avoid predators, and fecundity are much more
closely related to the size of the organism than to its age (Ebenman and Persson,
1988). Population models having size rather than age as their basic state variable
18
L. Persson and L.B. Crowder
may therefore be more appropriate in many situations when analyzing the
dynamics of stage-structured populations such as fish. Physiologically structured
models that have been developed during the past decade may be a useful tool here
(Metz et aI., 1988; DeAngelis and Gross, 1992; De Roos et al., 1992). The analysis
of popUlation dynamics in relation to habitat structure by using this modeling
approach has not yet been carried out. In a recent contribution, Persson et a1.
(submitted) showed that in planktivore-zooplankton systems, the population
dynamics for biologically realistic parameter values is characterized by recruiterdriven dynamics (regular cycles or quasiperiodic fluctuations). A shift from zooplankton feeding in the open water to feeding in the vegetation will decrease the
slope of the size scaling of the attack rate (Fig. 1.1 B) and increase the relative
competitive ability of small individuals versus large individuals (Fig. l.1C). This
will actually reinforce the tendency for recruiter-driven dynamics. Because population cycles are not commonly observed in fish populations except for obligate
planktivores (Hamrin and Persson, 1986), there must be some mechanism that
prevents large-amplitude population fluctuations. Vegetated habitats (refuges) in
combination with predator-induced restriction of habitat use of small vulnerable
size classes may be one way by which the population dynamics is stabilized as
larger-size classes are released from competition from recruits. To analyze this
potential stabilizing effect of vegetation on population and overall community
dynamics is a challenging task for future modeling research.
Acknowledgments. The research on which this review is partly based has been
supported by The Swedish Natural Science Research Council and the Swedish
Council for Forestry and Agricultural Sciences (to L. Persson) and by the U.S.
National Science Foundation and the University of North Carolina Sea Grant for
research on species interactions in submerged macrophytes and seagrass habitats.
(to L. Crowder). Valuable comments on the chapter were given by S. Diehl and an
anonymous reviewer, who are gratefully acknowledged.
References
Abrahams, M.Y.; Dill, L.M. A determination of the energetic equivalence of the risk of
predation. Ecology 70:999-1007; 1989.
Anderson, O. Optimal foraging by largemouth bass in structured environments. Ecology
65:851-861; 1984.
Bailey, K.M.; Houde, E.D. Predation on eggs and larvae of marine fishes and the recruitment problem. Adv. Mar. BioI. 25: 1-83; 1989.
Brabrand, A; Faafeng, B.A Habitat shift in roach (Rutilus rutilus) induced by pikeperch
(Stizostedion lucioperca) introduction: predation risk versus pelagic behaviour. Oecologia 95:38-56; 1993.0ecologia 95:38-46; 1993.
Brabrand, A.; Faafeng, B.A; Nilssen, J.P. Relative importance of phosphorus supply to
phytoplankton production: fish excretion verus external loading. Can. J. Fish. Aquat.
Sci. 47:364-372; 1990.
Breck, J.E.; Gitter, M.J. Effect of fish size on the reactive distance of bluegill (Lepomis
macrochirus) sunfish. Can. J. Fish. Aquat. Sci. 40:162-167; 1983.
Calder, W.A., III. Size, function and life history. Cambridge, MA: Harvard University
Press; 1984.
1. Fish-Habitat Interactions
19
Caraco, N.E; Cole, J.1.; Likens, G.E. New and recycled primary production in an oligotrophic lakes: insights for summer phosphorus dynamics. Limnol. Oceanogr. 37:590602; 1992.
Carpenter, S.R.; Kitchell, J.E The trophic cascade in lakes. Cambridge: Cambridge University Press; 1993.
Carpenter, S.R.; Frost, T.M.; Persson, L.; Power, M.; Soto, D. Freshwater ecosystems:
linkages of complexity and processes. In: Mooney, H. et aI., eds. Biodiversity and
ecosystem functions: a global perspective. New York: John Wiley and Sons; 1996:299325.
Christensen, B; Persson, L. Species specific antipredator behaviours: effects on prey choice
in different habitats. Behav. EcoI. Sociobiol. 32:1-9; 1993.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interactions between
bluegill and their prey. Ecology 63: 1802-1813; 1982.
DeAngelis, D.L.; Gross, L.J. Individual-based models and approaches in ecology-populations,
communities and ecosystems. New York: Chapman & Hall; 1992.
De Roos, AM.; Metz, J.AJ.; Diekmann, O. Studying the dynamics of structured population
models: a versatile technique and its application to Daphnia. Am. Nat. 139:123-147;
1992.
Diehl, S. Foraging efficiency of three freshwater fish: effects of structural complexity and
light. Oikos 53:207-214; 1988.
Diehl, S. Effects of habitat structure on resource availability, diet and growth of benthivorous
perch, Pereafluviatilis. Oikos 67: 403-414; 1993.
Diehl, S.; Ekl6v, P. Effects of piscivore-mediated habitat use on resources, diet and growth
of perch. Ecology 76: 1712-1726; 1995.
Ebenman, B.; Persson, L. Dynamics of size-structured populations-an overview. In:
Ebenman, B.; Persson, L., eds. Size-structured popUlations: ecology and evolution.
Heidelberg: Springer Verlag; 1988:3-9.
Ekl6v, P. Group foraging versus solitary foraging efficiency in piscivorous predators: the
perch, Perea fluviatilis. and pike, Esox lucius, patterns. Anim. Behav. 44:313-326;
1992.
Ekl6v, P.; Diehl, S. Piscivore efficiency and refuging prey: the importance of predator
search mode. Oecologia 98:344-353; 1994.
Ekl6v, P.; Hamrin, S.E Predator efficiency and prey selection: interactions between pike
Esox lucius, perch Perea fluviatilis and rudd Scardinius erythrophtlUllmus. Oikos 56:149156; 1989.
Ekl6v, P.; Persson, L. Species-specific antipredator capacities and prey refuges: interactions
between piscivorous perch (Perea /luviatilis) and juvenile perch and roach (Rutilus
rutilus). Behav. EcoI. Sociobiol. 37:169-178; 1995.
Fretwell, S.D.; Lucas, H.L. On territorial behaviour and other factors influencing habitat
distribution in birds. I. Theoretical development. Acta Biotheor. 19: 16-36; 1970.
Fuiman, L.A. Vulnerability of Atlantic herring larvae to predation by yearling herring. Mar.
Ecol. Prog. Ser. 51:291-299; 1989.
Fuiman, L.A. The interplay of ontogeny and scaling in the interactions of fish larvae and
their predators. J. Fish BioI. 4:55-79; 1994.
Fuiman, L.A; Magurran, A.E. Development of predator defences in fishes. Rev. Fish BioI.
Fish. 4:145-183; 1994.
Gasith, A Can littoral resources influence ecosystem processes in large, deep lakes? Verh.
Int. Verein. Theoret. Angew. LimnoI. 24:1073-1076; 1991.
Gerking, S.D. Feeding ecology offish. San Diego, CA: Academic Press; 1994.
Gilliam, J.E Habitat use and competitive bottlenecks in size-structured populations. Thesis,
Michigan State Univ., East Lansing; 1982.
Gilliam, J.E; Fraser, D.E Resource depletion and habitat segregation by competitors under
predation hazard. In: Ebenman, B.; Persson, L., eds. Size-structured populations: ecology and evolution. Heidelberg: Springer Verlag; 1988: 173-184.
20
L. Persson and L.B. Crowder
Greenberg, L.A.; Paszkowski, e.A; Tonn, W.M. Effects of prey composition and habitat
structure on foraging by two functionally distinct piscivores. Oikos 74:522-532; 1995.
Hall, D.J.; Werner, E.E.; Gilliam, J.P.; Mittelbach, G.G.; Howard, D.; Doner, C.G.; Dickerman, J.A; Stewart, AJ. Diel foraging behavior and prey selection in golden shiner. J.
Fish. Res. Bd. Can. 36:1929-1939; 1979.
Hamrin, S.P.; Persson, L. Asymmetrical competition between age classes as a factor causing
population oscillations in an obligate planktivorous fish. Oikos 47:223-232; 1986.
Hansson, L.-A.; Johansson, L.; Persson, L. Effects offish grazing on nutrient release and
succession of primary producers. Limno!. Oceanogr. 32:723-729; 1987.
He, X.; Kitchell, J.P. Direct and indirect effects of predation on a fish community: a whole
lake experiment. Trans. Am. Fish. Soc. 119:825-835; 1990.
He, X.; Wright, R.W. Piscivore-planktivore interactions in an experimental lake: population and community responses to manipulation. Can. J. Fish. Aquat. Sci. 49: 1176-1183;
1992.
Heck, H.L., Jf.; Crowder, L.B. Habitat structure and predator-prey interactions in vegetated
aquatic systems. In: Bell, S.S.; McCoy, E.D.; Mushinsky, H.R., eds. Habitat structurethe physical arrangement of objects in space. London: Chapman & Hall; 1991 :281-299.
Hugie, D.M.; Dill, L.M. Fish and game: a game theoretic approach to habitat selection by
predators and prey. J. Fish BioI. 45 (Supp!. A):151-169; 1994.
Lima, S.L.; Dill, L.M. Behavioural decisions made under risk of predation: a review and
prospectus. Can. J. Zoo I. 68:619-640; 1990.
Litvak, M.K.; Leggett, W.e. Age and size-selective predation on larval fishes: the bigger-isbetter hypothesis revisited. Mar. Eco!. Prog. Ser. 81: 13-24; 1992.
Lodge, D.M.; Barko, J.w.; Strayer, D.; Melack, J.M.; Mittelbach, G.G.; Howarth, R.W.;
Menge, B.; Titus, J.E. Spatial heterogeneity and habitat interactions in lake communities. In: Carpenter, S.R., ed. Complex interactions in lake communities. New York:
Springer-Verlag; 1988: 181-208.
Lundberg, S.; Persson, L. Optimal body size and resource density. J. Theor. Bio!. 164: 163180; 1993.
Main, K.L. Predator avoidance in seagrass meadows: prey behavior, microhabitat selection
and cryptic coloration. Ecology 68: 170-180; 1987.
Mather, M.E.; Vanni, MJ.; Wissing, T.E.; Davis, S.A; Schaus, M.H. Regeneration of
nitrogen and phosphorus by bluegill and gizzard shad: effect of feeding history. Can. J.
Fish. Aquat. Sci. 52:2327-2338; 1995.
McCauley, E.D.; Murdoch, W.W. Predator-prey dynamics in environments rich and poor in
nutrients. Nature 343:455-457; 1990.
Metz, J.A.J.; de Roos, AM.; van den Bosch, P. Population models incorporating physiological structure: a quick survey of the basic concepts and an application to size-structured
population dynamics in waterfleas. In: Ebenman, B.; Persson, L., eds. Size-structured
populations: ecology and evolution. Heidelberg: Springer-Verlag; 1988: 106-126.
Milinski, M. An evolutionarily stable strategy for sticklebacks. Z. Tierpsychol. 51:36-40;
1979.
Miller, T.J.; Crowder, L.B.; Rice, J.A; Marschall, E.A. Larval size and recruitment mechanisms
in fishes: toward a conceptual framework. Can. J. Fish. Aquat. Sci. 45: 1657-1670; 1988.
Mittelbach, G.G. Foraging efficiency and body size: a study of optimal diet and habitat use
by bluegills. Ecology 62:1370-1386; 1981.
Mittelbach, G.G. Predation-mediated habitat use: some consequences for species interactions. BioI. Fish. 16:159-169; 1986.
Mittelbach, G.G. Competition among refuging sunfishes and effects of fish density on
littoral zone invertebrates. Ecology 69:614-623; 1988.
Mittelbach, G.G.; Chesson, P.L. Predation risk: indirect effects on fish populations. In:
Kerfoot, W.e.; Sih, A, eds. Predation: direct and indirect impacts on aquatic communities. Hanover, NH: University Press of New England; 1987:537-555.
1. Fish-Habitat Interactions
21
Mittelbach, G.G.; Osenberg, C.W. Stage-structured interactions in bluegill: consequences
of adult resource variation. Ecology 74:2381-2394; 1993.
Mittelbach, G.G.; Osenberg, C.W.; Leibold, M.A Trophic relations and ontogenetic niche
shifts in aquatic systems. In: Ebenman, B.; Persson, L., eds. Size-structured populations:
ecology and evolution. Heidelberg: Springer-Verlag; 1988:203-218.
Naud, M.; Magnan, P. Diel onshore-offshore migrations in northern redbelly dace, Phoxinus eos
(Cope), in relation to prey distribution in a small oligotrophic lake. Can. J. Zool.
66:1249-1253; 1988.
Neill, W.E. Community responses to experimental nutrient perturbations in oligotrophic
lakes: the importance of bottlenecks in size-structured populations. In: Ebenman, B.;
Persson, L., eds. Size-structured populations: ecology and evolution. Heidelberg: SpringerVerlag; 1988:236---255.
Noakes, D.L.G.; Godin, J.-G.J. Ontogeny of behavior and concurrent developmental changes in sensory systems in teleost fishes. In: Hoar, W.S.; Randall, D.J., eds. Fish physiology. Vol. XIB. New York: Academic Press; 1988:345-395.
Oksanen, T.; Power, M.E.; Oksanen, L. Ideal free habitat selection and consumer-resource
dynamics. Am. Nat. 146:565-585; 1995.
Olson, M.H.; Mittelbach, G.G.; Osenberg, C.W. Competition between predator and prey:
resource-based mechanisms and implications for stage-structured dynamics. Ecology
76:1758-1771; 1995.
Osenberg, C.W.; Werner, E.E.; Mittelbach, G.G.; Hall, DJ. Growth patterns in bluegill
(Lepomis maeroehirus) and pumpkinseed (L. gibbosus) sunfish: environmental variation and the importance of ontogenetic niche shifts. Can. J. Fish. Aquat. Sci. 45:17-26;
1988.
Osenberg, c.w., Olson, M.H.; Mittelbach, G.G. 1994. Stage structure in fishes: resource
productivity and competition gradients. In: Stouder, D.J.; Fresh, K.L.; Feller, R.J., eds.
Theory and application in fish feeding ecology. Columbia: University of South Carolina
Press; 1994: 151-170.
Pepin, P.; Shears, T.H.; deLafontaine, Y. Significance of body size to the interaction
between a larval fish (Mallotus villosus) and a vertebrate predator (Gasterosteus
aeuleatus). Mar. Eea/. Prog. Ser. 81:1-12; 1992.
Persson, L. Effects of reduced interspecific competition on resource utilization in perch
(Pereafluviatilis). Ecology 67:355-364;1986.
Persson, L. Asymmetries in competitive and predatory interactions in fish populations
evolution. In: Ebenman, B.; Persson, L., eds. Size-structured populations: ecology and
evolution. Heidelberg: Springer-Verlag; 1988:203-218.
Persson, L. Behavioral response to predators reverses the outcome of competition between
prey species. Behav. Ecol. Sociobiol. 28:101-105; 1991.
Persson, L. Predator-mediated competition in prey refuges: the importance of habitat
dependent prey resources. Oikos 68: 12-22; 1993.
Persson, L.; Ek16v, P. Prey refuges affecting interactions between piscivorous perch (Perea
fluviatilis) and juvenile perch and roach (Rutilus rutilus). Ecology 76:70-81; 1995.
Persson, L.; Greenberg, L.A Juvenile competitive bottlenecks: the perch (Perea fluviatilis}roach (Rutilus rutilus) interaction. Ecology 71 :44-56; 1990.
Persson, L.; Leonardsson, K.; Gyllenberg, M.; De Roos, A; Christensen, B. Ontogenetic
scaling and the dynamics of a size-structured consumer-resource model. Theor. Pop.
BioI. (submitted).
Persson, L.; Diehl, S.; Johansson, L.; Andersson, G.; Harnrin, S.F. Shifts in fish communities along the productivity gradient of temperate lakes-patterns and the importance of size-structured interactions. J. Fish BioI. 38:281-293; 1991.
Persson, L.; Johansson, L.; Diehl, S.; Andersson, G.; Hamrin, S.F. Trophic interactions in
temperate lake ecosystems-a test of food chain theory. Am. Nat. 140:59-84; 1992.
Peters, R.H. The ecological implications of body size. Cambridge: Cambridge University
Press; 1983.
22
L. Persson and L.B. Crowder
Prejs, A. Herbivory by temperate freshwater fishes and its consequences. Environ. BioI.
Fish. 10:281-296; 1984.
Reinertsen, H.; Jensen, A.; Langeland, A.; Olson, Y. Algal competition for phosphorus: the
influence of zooplankton and fish. Can. J. Fish. Aquat. Sci. 43:1135-1141; 1986.
Ryer, C.H. Studies on the foraging ecology of pipefish: prey selection and habitat selection.
Dissertation, College of William and Mary, Williamsburg, VA; 1987.
Ryer, C.H. Pipefish foraging and the effects of altered habitat complexity. Mar. Ecol. Progr.
Ser. 48:37-45; 1988.
Savino, J.E; Stein, R.A. Predator-prey interaction between largemouth bass and bluegills
as influenced by simulated, submersed vegetation. Trans. Am. Fish. Soc. 111:255-266;
1982.
Savino, J.E; Stein, R.A. Behavioural interactions between fish predators and their prey:
effects of plant density. Anim. Behav. 37: 11-321; 1989a.
Savino, J.E; Stein, R.A. Behavior of fish predators and their prey: habitat choice between
open water and dense vegetation. Environ. BioI. Fish. 24:287-293; 1989b.
Schindler, D.E. Nutrient regeneration by sockeye salmon (Oncorhynchus nerka) fry and
subsequent effects on zooplankton and phytoplankton. Can. J. Fish. Aquat. Sci. 49:
2498-2506; 1992.
Schindler, D.E.; Kitchell, IE; He, X.; Carpenter, S.R.; Hodgson, J.R.; Cottingham, KL.
Food web structure and phosphorus cycling in lakes. Trans. Am. Fish. Soc. 122:756772; 1993.
Schindler, D.E.; Carpenter, S.R.; Cottingham, KL.; He, X.; Hodgson, J.R.; Kitchell, J.E;
Soranno, P.A. Food web structure and littoral zone coupling to pelagic trophic cascades.
In: Polis, G.A.; Winemiller, KO., eds. Food webs: integration of patterns and dynamics.
New York: Chapman & Hall; 1996:95-105.
Sih, A. The trophic level ideal free distributions: a game theory approach to understanding
the predator-prey behavioral response race. In: Dugatkin, L.A.; Reeve, H.K, eds.
Advances in game theory and the study of animal behavior. Oxford: Oxford University
Press (in press).
Stein, R.A.; Threlkeld, S.T.; Sandgren, C.D.; Sprules, W.G.; Persson, L.; Neill, w.E.;
Werner, E.E.; Dodson, S.I. Size-structured interactions in lake communities. In: Carpenter, S.R., ed. Complex interactions in lake communities. Heidelberg: Springer-Verlag; 1988:161-179.
Tonn, W.M.; Paszkowski, c.A.; Holopainen, 1.1. Piscivory and recruitment: mechanisms
structuring prey populations in small lakes. Ecology 73:951-958; 1992.
Tripet, E; Perrin, N. Size-dependent predation by Dugesia lugubris (Turbellaria) on Physa
acuta (Gastropoda): experiments and model. Funct. Ecol. 8:458-463; 1994.
Turner, A.M.; Mittelbach, G.G. Predator avoidance and community structure: interactions
among piscivores, plankti vores, and plankton. Ecology 71 :2241-2254; 1990.
Vanni, MJ. Nutrient transport and recycling by consumers in lake food webs: implications
for algal communities. In: Polis, G.A.; Winemiller, KO., eds. Food webs: integration of
patterns and dynamics. New York: Chapman & Hall; 1996:81-95.
Vanni, M.J.; Findlay, D.L. Trophic cascades and phytoplankton community structure.
Ecology 71:921-937; 1990.
Vanni, M.J.; Layne, C.D.; Amott, S.E. "Top-down" trophic interactions in lakes: effects of
fish on nutrient dynamics. Ecology 78:21-40.
Vanni, M.J.; Layne, C.D. Nutrient recycling and herbivory as mechanisms in the "top
down" effect of fish on phytoplankton in lakes. Ecology 78: 1-20.
Webb, P.W. Hydromechanics: nonscombrid fish. In Hoar, W.S.; Randall, D.J., eds. Fish
physiology. Vol. VII. Locomotion. New York: Academic Press; 1978:190-237.
Webb, P.w. Body and fin form and strike tactics of four teleost predators attacking fathead
minnow Pimephales promelas prey. Can J. Fish. Aquat. Sci. 41:157-165; 1984.
Werner, E.E. Species packing and niche complementarity in three sunfishes. Am. Nat.
11:553-578; 1977.
I. Fish-Habitat Interactions
23
Werner, E.E. Species interactions in freshwater fish communities. In: Diamond, J.; Case,
TJ., eds. Community ecology. New York: Harper & Row; 1986:344-358.
Werner, E.E. Size, scaling and the evolution of life cycles. In: Ebenman, B.; Persson, L.,
eds. Size-structured populations: ecology and evolution. Heidelberg: Springer-Verlag;
1988 :60-81.
Werner, E.E. Ontogenetic scaling of competitive relations: size-dependent effects and
responses in two Anuran larvae. Ecology 75: 197-231; 1994.
Werner, E.E.; Gilliam, J.E The ontogenetic niche and species interactions in size-structured
populations. Annu. Rev. Ecol. Syst. 15:393-426; 1984.
Werner, E.E.; Hall, D.J. Foraging efficiency and habitat switching in competing sunfishes.
Ecology 60:256-264; 1979.
Werner, E.E.; Hall, D.J. Ontogenetic habitat shifts in bluegill: the foraging rate-predation
risk trade-off. Ecology 69:1352-1366; 1988.
Werner, E.E.; Gilliam, J.E; Hall, D.J.; Mittelbach, G.G. An experimental test of the effects
of predation risk on habitat use in fish. Ecology 64: 1540-1548; 1983.
Wetzel, RG. The role of the littoral zone and detritus in lake metabolism. Arch. Hydrobio!.
13:145-161; 1979.
Wetzel, RG. Land-water interfaces: metabolic and limnological indicators. Verh. Int.
Verein. Theor. Angew. Limno!. 24:6-24; 1990.
Wilbur, H.M. Interactions between growing predators and growing prey. In: Ebenman, B.;
Persson, L., eds. Size-structured popUlations: ecology and evolution. Heidelberg: SpringerVerlag; 1988:157-172.
Wilson, D.S. The adequacy of body size as a niche difference. Am. Nat. 109:769-784;
1975.
Winfield, I.J. The influence of simulated aquatic macrophytes on the zooplankton consumption rate of juvenile roach, Rutilus rutilus, rudd, Scardinius erythrophthalmus, and
perch, Percafluviatilis. J. Fish Bio!. 29:37-48; 1986.
2. Influence of Submerged Macrophytes on
Trophic Interactions Among
Fish and Macroinvertebrates
Sebastian Diehl and Ryszard Komij6w
Introduction
Lentic macroinvertebrates have received comparatively little interest from lake
managers and researchers for two reasons. First, with few exceptions (e.g., the
control of mosquitos, the provision of forage for waterfowl), there is no public
interest in managing ponds and lakes for macroinvertebrates per se. Macroinvertebrates are not a directly harvestable resource, such as fish, and are rarely
perceived as a nuisance, such as excessively growing planktonic algae or macrophytes. Second, macroinvertebrates are impractical study organisms. They are
taxonomically and functionally diverse, seasonally and geographically highly
variable, strongly patchy in their local distributions, and tedious to sample and
sort. This combination of high natural variability, high sampling effort, and low
sampling precision makes it notoriously difficult to distinguish pattern from error
and noise in estimates of macroinvertebrate responses to comparative or experimental gradients. Furthermore, many macroinvertebrates cannot be observed
in their natural environments and are very difficult to manipulate in the field,
which limits the approaches to study their behaviors and impacts on other organisms.
Despite these constraints to their study, macroinvertebrates perform important
ecosystem and community functions in lakes, many of which are mediated
through the interactions between macroinvertebrates and fish. For example, the
consumption of sediment-living, detritivorous macroinvertebrates by fish con24
2. Fish-Macroinvertebrate Interactions
E
Q)
.§
·s
25
Piscivorous
Fish
U
Q)
a::
Submerged
Macrophytes
Macroinvertebrates
Figure 2.1. Effects of submerged macrophytes on fish-macroinvertebrate interactions
considered in this chapter (thick solid arrows). Thin solid arrows paralleled by solid lines
with vortices represent trophic interactions. Broken arrow represents recruitment processes.
stitutes a significant pathway of phosphorus recycling from sediments to the water
column (Andersson et aI., 1988). Macroinvertebrates are a major food source for
many fish species during at least some ontogenetic stages, and the availability of
macroinvertebrate resources has important consequences for the abundances and
population size structures of various fish species (Persson, 1988; Osenberg et al.,
1992; Olson et aI., 1995). Conversely, fish predation acts as a major selective
mortality agent on macroinvertebrates, with repercussions for the abundance, species
composition, and size structure of rnacroinvertebrates (Erikson et al., 1980; Morin,
1984; Osenberg and Mittelbach, 1989; Diehl, 1995; Kornij6w, 1997).
The abundance of various species of submerged macrophytes in the littoral
zone of lakes influences fish-macroinvertebrate interactions in a variety of ways,
affecting processes on a wide range of spatial and temporal scales (Fig. 2.1). For
example, submerged macrophytes increase the diversity of habitats and resources
for macroinvertebrates and, thus, the basis of secondary production; they reduce
the susceptibility of macroinvertebrates to fish predators; they reduce the vulnerability of prey fish to piscivores (Diehl, 1988; Komij6w et al., 1990; Ekl6v and
Diehl, 1994). All of this affects the habitat choice, survival, and growth of macroinvertebrate-feeding fish, ultimately feeding back on the population dynamics of
fish and the predation pressure by fish on macroinvertebrates and other components of the lake food web (Persson et al., 1992; Mitte1bach and Osenberg,
1993; Diehl and EklOv, 1995).
Experimental studies of fish-macroinvertebrate interactions have been largely
restricted to small spatial and short temporal scales. The understanding of the
large-scale, long-term effects of submerged vegetation on fish-macroinvertebrate
interaction is still sketchy and will require increased efforts to identify and quantify feedback mechanisms among processes in littoral, benthic and pelagic habitats
(Lodge et al., 1988; Osenberg et aI., 1994; Persson et al., 1997). In this chapter, we
26
S. Diehl and R Kornij6w
review experimental and comparative studies of the influence of submerged macrophytes on trophic interactions among fish and macroinvertebrates (Fig. 2.1). We
first review empirical relationships among submerged vegetation and the production, distribution, and abundance of macroinvertebrates in the littoral zone of
lakes. Based on data from within-season experiments, we then discuss behavioral
and trophic interactions among fish and macroinvertebrates and how these are
affected by submerged macrophytes. Finally, we explore the effects of submerged
vegetation on the long-term dynamics of fish and macroinvertebrates in the context of simple predator-prey models and discuss possible complications arising
from ontogenetic, size-related habitat and diet shifts of fish. We conclude with a
few suggestions for future research.
Relationships Between Submerged Macrophytes and
Littoral Macroinvertebrates
Benthic and Epiphytic Macroinvertebrates
Littoral macroinvertebrates can be categorized according to the microhabitats in
which they occur as being benthic (i.e., dwelling the bottom sediments), and
epiphytic (i.e., associated with macrophytes). Benthic and epiphytic macroinvertebrates often belong to the same classes, orders, or families but rarely to the same
genera (Pieczynski, 1977; Komij6w et aI., 1990; Komij6w and Kairesalo, 1994a).
Most littoral macroinvertebrates are typical of either benthic or epiphytic faunas,
but there are some widespread opportunistic species such as Asellus aquatic us
(Isopoda), Helobdella stagnalis (Hirudinea), and Lymnaea peregra (Gastropoda)
that readily inhabit and move between both benthic and epiphytic habitats. Benthic
and epiphytic macroinvertebrates differ with respect to their seasonal dynamics, food
sources, and predators and should therefore be distinguished both conceptually
and empiricall y (Pieczynski, 1977; Pardue and Webb, 1985; Komij6w et aI., 1990;
see Kajak et aI., 1965; Mittelbach, 1981b; Downing, 1984; Kornij6w and Kairesalo, 1994b, for recommendations on sampling techniques). To avoid confusion, in
this chapter we generally use the inclusive term macro invertebrates rather than
benthos whenever it is either not necessary or not possible (based on the data
reported in the primary literature) to distinguish between benthic and epiphytic
macroinvertebrates.
Resource Use, Distribution, and Abundance of Macroinvertebrates in
Relation to Submerged Macrophytes
Submerged macrophytes affect abiotic variables such as light, temperature, and
oxygen concentration and provide additional habitat and resources for macroinvertebrates (Carpenter and Lodge, 1986; Kornij6w et aI., 1990; Lillie and Budd,
1992; Kornij6w and Kairesal0, 1994a). For example, many herbivores benefit
from the additional substrate area provided by submerged macrophytes to periphytic
27
2. Fish-Macroinvertebrate Interactions
N
E 20000
.s--
A
OJ
rJl
rJl
!1l
15000
E
0
C5
2 10000
•
0 Benthic
~
.0
Q)
t::
Q)
>
c
"eu
!1l
:::2:
Epiphytic
5000
0
Low Intermediate High
No
Macrophyte Density
rJl
B
100
Q)
~
.0
Q)
t::
80
>
c
'eu
60
:::2:
40
!1l
u
~
.c
0'0.
20
~
0
0
•
••
•
•II
•
• • •
Q)
•
••
w
0
1
2
3
Macrophyte Biomass (kg/m2)
Figure 2.2. Relationships between macrophyte density and (A) the wet biomass of benthic
and epiphytic macroinvertebrates in Lake Paajiirvi (with 95% confidence intervals), and
(B) the percentage of total macro invertebrate biomass composed of epiphytic macroinvertebrates in Lake Memphremagog. The fitted curve is described by y = 100axl(ax + b), with
a = 0.093, b = 0.102, ? = 0.87. (Data from (A) Kornij6w and Kairesalo, 1994a, and
(B) Rasmussen, 1988.)
algae (Jones et aI., this volume, Chapter 4; Bronmark and Vermaat, this volume,
Chapter 3). Fresh tissue of macrophytes can also be important in the diets of some
herbivores (Komij6w, 1996; Lodge et al., this volume, Chapter 8). After senescence and death, epiphytic algae and macrophytes become available as a food
source to shredders and deposit feeders (Suren and Lake, 1989; Komij6w et aI.,
1995). Through these various pathways, submerged macrophytes have a strong,
positive effect on the resource base of epiphytic and benthic macroinvertebrates,
with the possible exception of some benthic collector-suspension feeders (e.g.,
chironomid larvae of the genus Chironomus) that may experience a reduced
28
s. Diehl and R Kornij6w
supply of fresh and dead planktonic algae under the cover of submerged macrophytes (Kornij6w and Moss, this ~olume, Chapter 12).
The taxonomic diversity and density of epiphytic macro invertebrates are related to the growth forms of the macrophytes on which they live, but generally
there is a strong positive relationship between the abundances of submerged
macrophytes and epiphytic macro invertebrates (Figs. 2.2 and 2.6A; Cyr and
Downing, 1988a,b; Jeffries, 1992; Lillie and Budd, 1992). Similarly, benthic
macroinvertebrates are often more abundant in the sediments of vegetated areas
compared with open areas (Fig. 2.2; Prejs, 1976; Pardue and Webb, 1985; Schramm
and Jirka, 1989; Kornij6w et aI., 1990, but see the opposite pattern for benthos
dominated by chironomids in Blindow et aI., 1993). The relative contribution of
epiphytic macroinvertebrates to total macro invertebrates increases with increasing
vegetation density, and in dense stands of submerged macrophytes, epiphytic
macroinvertebrates are usually more abundant than benthic macroinvertebrates
(Fig. 2.2; Pieczynski, 1977; Mittelbach, 1981b; Schramm and Jirka, 1989; Kornijow et al., 1990). The positive effects of submerged vegetation are usually
smaller on the biomass than on the abundance of macro invertebrates, because
epiphytic macroinvertebrates are, on average, smaller than benthic macroinvertebrates (Mittelbach, 1981b; Diehl, 1992; Rasmussen, 1993; Kornijow and Kairesalo, 1994a; Diehl and EklOv, 1995, but see Blindow et aI., 1993, for a counter
example). In absolute numbers or biomass, however, large, profitable (for fish)
macroinvertebrates are more abundant in vegetated than unvegetated littoral
habitats (Diehl, 1993a; Blindow et aI., 1993).
Most of the data establishing the positive relationship between macro invertebrates and submerged macrophytes were collected in lakes and ponds containing
fish. The relationship is therefore likely to be a result not only of the positive
impact of submerged macrophytes on the productivity of macroinvertebrates, but
also of the complex influence of vegetation on the interactions between macroinvertebrates and their fish predators (see below).
Short-Term Interactions Between Fish and Macroinvertebrates
Behavioral Interactions
The interactions between fish and macroinvertebrates have been studied in the
laboratory and in field experiments on relatively small spatial scales (laboratory: < 1 m 2; field: < 10 m2) and over relatively short temporal scales (laboratory:
hours to days; field: weeks to months). These studies show four broad patterns, all
of which suggest that fish-macro invertebrate interactions may be strongly influenced by the behavior of fish and macroinvertebrates.
First, many macroinvertebrates adjust their behavior in response to fish by
seeking refuge and/or decreasing risky activities (Huang and Sih, 1990; McPeek,
1990b; Ball and Baker, 1995). Such risk-sensitive prey behavior can have profound consequences for the population dynamics of both prey and predators
(Abrams, 1992; Werner, 1992).
29
2. Fish-Macroinvertebrate Interactions
Figure 2.3. Relationships between fish
density treatments and (A) macroinvertebrate abundance, and (B) chironomid
abundance at the end of 3-month field
enclosure experiments. Submerged macrophytes were dense in A and absent in
B. Fish were bluegill plus pumpkinseed
sunfish (A) and perch (B). The fitted curves are described by y = a + b exp(-cx).
For A: a = 363, b = 867, c = 0.973, ? =
0.59; for B: a = 954, b = 6.05*108, C =
12.6, ? = 0.84. (Data from [A] Mittelbach, 1988, and [B] Diehl, 1995.)
1\1
E
.......
~
Q)
u
c
•
1000
.~.
c
:J
•
.. ~
ctl
'0
.0
ctl
A
1500
0
500
>.
Q)
....
a...
1\1
--
E
0
4000
ci
~
Q)
u
c
ctl
'0
2000
C
:J
.0
ctl
,.---,-
0.0
a...
0
i
2.0
i
•
i
4.0
3.0
B
•
\~
>.
....
Q)
1.0
0
•
•
•
•
•
•
•
0.17
0.33
0.5
0.67
Fish density [No.lm2]
Second, predation by fish is selective. Generally, large macroinvertebrate taxa
are more vulnerable to fish predation than small taxa (Crowder and Cooper, 1982;
Morin, 1984; McPeek, 1990a; Diehl, 1992). This pattern appears to hold also on
larger spatial and longer temporal scales as is indicated by comparative studies of
lake systems with and without fish (Erikson et al., 1980; Bendell and McNicol,
1987; Evans, 1989; Bradford et al., in press). Exceptions are taxa with morphological or behavioral defenses such as snails with size refuges, odonates with
cryptic behavior, or sediment-dwelling chironomids with refuges in deeper sediment layers (Osenberg and Mittelbach, 1989; McPeek, 1990b; Macchiusi and
Baker, 1991; Kornij6w, in press; Kornij6w and Moss, this volume, Chapter 12).
Third, in studies spanning over more than two fish density treatments, negative
effects of fish on the abundance or biomass of macroinve11ebrates are usually
strongest in the lowest range of fish densities, and the relationship between prey
abundance and fish density can often be approximated by an t:xponentially declining curve (Fig. 2.3). In part, this may simply reflect the time scale of the studies.
In a short-term study, if total prey consumption were directly proportional to
predator and prey abundances, prey decline across predator densities would be
expected to follow a simple exponential model (see Diehl, 1995; Osenberg and
Mittelbach, 1996, for detailed discussions). In many studies, however, macroinvertebrate abundances appear to asymptote at nonzero values (Fig. 2.3), and
behavioral processes are likely to be responsible for these asymptotes. For example, because the negative effects of fish are strongest on preferred taxa and
30
:§
..c:
S. Diehl and R Kornij6w
1.5
A
(/)
;;
....
1.0
<D
0.5
<D
c..
(/)
a)
<D
....
()
.~
/
0.0
a)
:::2:
:§
11
7
10
(/)
(/)
3
-0.5
0
400
800
1200
5
1
=No vag .• low parch
~
4
/
2
c..
3
..c:
Qj
<D
(/)
a)
~
2
()
.~
(/)
(/)
a)
:::2:
0
3
B
=Vag., low perch
=No vag., high perch
~
4
=Veg., high parch
0 1000 2000 3000 4000
Macroinvertebrates [mg/m2]
Figure 2.4. Relationships between rnacroinvertebrate prey density and the average
individual growth rates of fish in field enclosures. (A) Linear regression statistics:
Y =0.023 + 0'(X)094x, ,:z. =0.43. Data points
in A are identified by their fish density treatments (number of bluegill plus pumpkinseed sunfishl3 m2). Note that, at a given
prey density, fish growth is consistently Iower at higher fish densities. Fish density explains 39% of the variation in the residuals
of the shown relationship (residuals =0.30.6*fish density, ,:z. = 0.39, P = .055). (B)
Lines 2 to 4 are linear regressions fitted to
replicates of treatments varying in vegetation density (submerged macrophytes absent
and present) and fish density (0.33 and 1
perchlm2). In the treatment depicted by line
I, there was no relationship between prey
density and fish growth, and the mean
growth of fish is shown over the range of
prey densities encountered. (Data from [A]
Mittelbach, 1988, and [B] Diehl, 1993a.)
large size classes of macroinvertebrates, these prey items are frequently reduced
already at low fish densities, and no further reduction may occur at higher fish
densities. Similarly, because many macroinvertebrates adjust their behaviors to the
risk of predation by fish, fish may be unable to reduce their prey below certain
levels at which all prey make themselves unavailable to fish. Behavioral interactions among macroinvertebrates and fish may, in part, explain, why studies in
which the lowest fish density treatments are nonzero often report only relatively
subtle effects of fish density on macroinvertebrates (e.g., effects on selected taxa
or on the size structure of the macroinvertebrate assemblage) (Persson and Greenberg, 1990; Bergman and Greenberg, 1994; Olson et al., 1995). Furthermore,
nonzero asymptotes of macroinvertebrate density (Fig. 2.3) suggest that the degree
to which macroinvertebrates can make themselves unavailable to fish may be
limited by the availability of physical refuges.
Finally, the feeding and individual growth rates of many fish species are
positively related to the abundance of macroinvertebrates (Fig. 2.4). Different fish
species may, however, vary considerably in their efficiencies at exploiting macroinvertebrate prey (Diehl, 1988; Persson, 1988). Furthermore, at comparable
macroinvertebrate densities, individual fish grow less at higher own densities
(Fig. 2.4). This suggests that macroinvertebrate-feeding fish have negative effects
on their own growth rates, which are stronger than would be expected if pure
31
2. Fish-Macroinvertebrate Interactions
exploitative competition for macroinvertebrates were the only mechanism of
density dependence. Again, the mechanisms that account for this additional
density dependence are possibly behavioral (e.g., interference among fish
and/or reduced activity and increased hiding of macroinvertebrates at higher
fish densities).
Omnivory
Macroinvertebrate-feeding fish are omnivores (i.e., they consume prey from more
than one trophic level) (Diehl, 1993b). Omnivory complicates the trophic relationships among fish and macro invertebrates, because fish and invertebrate predators
engage in both competitive and consumer-resource interactions. For example, fish
could have an indirect positive effect on the abundance of nonpredatory macroinvertebrates by numerically releasing them from invertebrate predators (Diehl,
1992; Polis and Holt, 1992; Hill and Lodge, 1995; Prejs et aI., 1977). Because
most invertebrate predators are bigger than most nonpredatory invertebrates,
preferential predation by fish on large macroinvertebrates provides another
potential mechanism for an indirect positive effect of fish on nonpredatory
macroinvertebrates. Decreased predation by fish on (smaller) nonpredatory macroinvertebrates in the presence of increased numbers of (larger) predatory macroinvertebrates has, in fact, been demonstrated (Fig. 2.5). Compared with fishless
treatments, however, fish very rarely seem to have overall positive effects on
nonpredatory macroinvertebrates and always have negative effects on predatory
macroinvertebrates over the time scales of typical enclosure studies (Fig. 2.5,
Diehl, 1993b). Unless additional mechanisms stabilize the intermediate predator (Diehl, 1993b), such a situation should not be sustainable; that is, predatory macroinvertebrates can only coexist with fish, if they, in the absence of
fish, suppress nonpredatory macro invertebrates below the levels at the threetrophic level equilibrium with fish present (R*N < R**; Fig. 2.7C, Polis and
Holt, 1992; Holt and Polis, 1997).
E
Figure 2.5. Relationship between density
of chironomids at the end of a 3-month
field enclosure experiment and initial density of Sialis lutaria (a predatory macroinvertebrate) in the presence (data points
and solid line) and in the absence of omnivorous perch (dashed line; only the mean
is shown, because there was no relationship between the densities of chironomids
and Sialis). (Data from Diehl, 1995.)
--o
4000
~
(f)
"0
'E
2000
•
0
c
g
..c:
()
--------:•
0
0
40
80
120
Sia/is [No./m2]
160
32
s. Diehl and R Kornij6w
Effects of Submerged Macrophytes on Short-Term Interactions
Between Fish and Macroinvertebrates
Numerous laboratory experiments have demonstrated that the rates at which most
fish species encounter and attack prey decline with increasing density of artificial
or natural macrophytes (Diehl, 1988; Nelson and Bonsdorff, 1990, and references
therein). This has two important consequences for the interactions between fish
and macroinvertebrates. First, at a given prey density, individual growth rates of
macroinvertebrate-feeding fish are negatively affected by vegetation density (see
lines 1 vs. 2 and 3 vs. 4 in Fig. 2.4B). In other words, for individual fish to attain
the same feeding and growth rates, prey densities have to be higher the higher the
density of vegetation. Second, at a given fish density, survival and, subsequently,
the density of macroinvertebrates are higher in densely compared with sparsely
vegetated habitats (see below).
The negative effects of vegetation on the encounter rates of fish and the positive
effects of vegetation on the standing stock and production of macroinvertebrates
make it impossible to derive a general relationship between the growth rates of
fish and the density of submerged vegetation. Positive, negative, and neutral
effects of increased use of high-density vegetation patches on the individual
growth of invertebrate-feeding fish have been reported (Crowder and Cooper,
1982; Turner and Mittelbach, 1990; Diehl, 1993a; Diehl and EklOv, 1995; Persson
and Ekl5v, 1995).
It is likely that the relationship between the individual growth of fish and the
density of submerged vegetation depends on the morphology, size, and sensory
capacities of the fish as well as on the growth form of the macrophytes. For
example, primarily visually searching fish such as Eurasian perch (Percajluviatilis),
bluegill sunfish (Lepomis macrochirus), and largemouth bass (Micropterus salmoides)
are able to feed on epiphytic macroinvertebrates, which constitute the bulk of macroinvertebrate resources in vegetated habitats (Fig. 2.2B). In such species, maximum
individual growth rates may be achieved at some intermediate level of vegetation
density (Crowder and Cooper, 1982; Wiley et aI., 1984; Savino et al., 1992). By
contrast, primarily tactile bottom feeders such as bream (Abramis brama) are
likely to be uniformly negatively affected by increasing vegetation density (Diehl,
1988).
Effects of Vegetation on the Long-Term Dynamics of
Fish and Macroinvertebrates
Short-term and small-scale enclosure studies are insufficient to understand and
predict the long-term effects of submerged vegetation on fish-macroinvertebrate
interactions in whole-lake systems. The nonequilibrium nature of enclosure
studies raises issues of feedback of fish predation on the long-term population
dynamics of macroinvertebrates (including reproduction and dispersal) and of
33
2. Fish-Macroinvertebrate Interactions
Figure 2.6. Relationship among the average
wet biomasses of macrophytes, macroinvertebrates, and fish in the littoral zones of ten
southern Quebec lakes. Linear regression
statistics are: (A) y = 12.2 + 0.015x,
=
0.42, P = .057; (B): y = 2.2 + 0.0051x, ? =
0.40, P = .066. One lake (open triangles)
was excluded from the analyses, because it
had high biomasses (>12 glm2) of two fish
species (Moxostoma anisurum and M. valenciennesi) that were absent from all other
lakes. (Data from Pierce et aI., 1994.)
r
C\J
70
-.9
60
en
Q)
50
E
~
40
Q)
30
>
c::
'0
.....
Q)
20
CI:l
0
.0
1::
u
~
/
"/
•
10
0
I::.
1000
•
2000
3000
30
•
C\J
--
E
20
I::.
.c:
en
u:::
•
10
4000
B
•
•
.9
A
•
•
0
0
1000
2000
3000
4000
Macrophytes (g/m2)
feedback of the abundances of macroinvertebrates and submerged macrophytes on
the population dynamics of macroinvertebrate-feeding fish. An obvious and important question is: What are the effects of submerged macrophytes on the standing stock and production of macroinvertebrates and macroinvertebrate-feeding
fish in natural systems? Data to address the question at the lake scale are so far
scarce, but they suggest that submerged macrophyte density may have a positive
effect on the abundances of both macroinvertebrates and fish (Fig. 2.6; Blindow et
aI., 1993). In the absence of an extensive body of empirical data at the whole-lake
scale, our approach here is to derive plausible expectations on the long-term,
large-scale dynamics of fish and macroinvertebrates based on knowledge gained
at smaller spatial and shorter temporal scales. We first explore potential effects of
submerged vegetation on the population dynamics of macroinvertebrates and fish
in the context of simple predator-prey models, incorporating mechanisms observed in short-term experiments as model assumptions. The qualitative predictions derived from this exercise will then serve as reference points in the
subsequent discussion of the more complex (and more realistic) situation in which
fish populations are size-structured. It will become obvious that the vegetationmacro invertebrate-fish interaction cannot be understood in isolation from other
components of the lake food web and from processes that link the littoral zone to
benthic and pelagic habitats (see also Lodge et al., 1988; Persson and Crowder,
this volume, Chapter 1).
S. Diehl and R Kornij6w
34
Expectations from Predator-Prey Theory
Considerations based on simple predator-prey theory suggest that increasing the
density of submerged vegetation may often have positive effects on the long-term
abundances of both macro invertebrates and their fish predators and can stabilize
the dynamic interaction between the two (Fig. 2.7 A). Based on the evidence from
short-term experiments, we can assume that increasing vegetation density increases macroinvertebrate productivity and reduces the search efficiency of macroinvertebrate-feeding fish. In a graphical analysis of a simple predator-prey model
Figure 2.7. Potential effects of vegetation on equilibrium densities of fish and macroinvertbrates. (A) Examples of macroinvertebrate (R) and fish (P) isoclines based on a
differential equation predator-prey model assuming logistic prey growth with intrinsic
growth rate r and carrying capacity K. a type II functional response of the predator f(R). a
constant conversion efficiency c of prey into predators, and a constant mortality rate m of the
predator. The dynamical equations are
dR
R
dt
= rR(l - K)
- f(R)P
and
dP
dt
= cf(R)P -
mP
·th f(
WI
aR
R) = I + abR
where a is the predator's search rate and b its handling time per prey. Isoclines connect all
possible combinations of predator and prey densities at which the instantaneous rates of
population change of either the prey (dRldt) or the predator (dPldt) are zero. Prey densities
increase (decrease) anywhere below (above) the prey isocline. Predator densities increase
(decrease) anywhere to the left (right) of the predator isocline. Intersections of the predator
and prey isoclines denote local equilibria. Equilibria are locally stable when the prey
isocline has a negative slope at the intersection with the predator isocline. Relative stability
(measured as resilience [i.e., the reciprocal of the return time to equilibrium]) increases with
the steepness of the negative slope of the prey isocline at equilibrium. Hatched lines and
equilibrium I represent the baseline case. Equilibrium 2 illustrates a case in which an
increase in vegetation density solely decreases the predator's search rate a. Equilibrium 3
illustrates a case in which an increase in vegetation density additionally increases the
carrying capacity K of the prey. (B) Prey production at eqUilibrium in cases I and 2 of A. The
heights of the vertical lines represent the harvestable prey production. (C) Example of
isoclines of predatory (N) and nonpredatory (R) macroinvertebrates and fish (P) based on a
differential equation food chain model assuming omnivory by the fish, logistic growth of
nonpredatory macroinvertebrates, linear functional responses, and constant conversion efficiencies of resources into consumers. The intersection of the three planes denotes the
equilibrium with all three trophic levels present. K denotes the carrying capacity of R in the
absence of Nand P, R*N and R*p the equilibria of R with only N or only P present,
respectively, and R** the equilibrium of R with all three trophic levels present. Note that the
origin is in the lower front comer and that the isocline of R stretches from near the upper
front corner down and back into the phase space. Increasing K will move the isocline of R
up and to the right and push it back. The three-trophic level equilibrium (R**, N**, P**)has
to fall onto the intersection of the isoclines of N and P, which slants downward (toward
lower N) with increasing P. Therefore, an increase in K will cause both R** and P** to
increase but N** to decrease.
35
2. Fish-Macroinvertebrate Interactions
dP/dt = 0
"
Q)
~a:-
A
3
.0Q)J::
t::
Q)u.
,,
o
.... "0
-----t----_
.~
~C)
.- c:
:1
~ ~
~-
Macroinvertebrates (R) low K
high K
8
2
Macroinvertebrates (R) low K
36
S. Diehl and R Kornij6w
(Fig. 2.7A), reducing the predator's search efficiency moves the predator isocline
to the right (because higher prey densities are required for the predators to attain
the feeding rates at which predator production exactly offsets mortality) and the
prey isocline up (because more predators are needed to consume a given number
of prey). Increasing prey productivity moves the prey isocline further up and also
to the right. Consequently, with an increase in vegetation density, prey will always
equilibrate at a higher density, whereas the effects on the equilibrium density of
predators and on the local stability of the equilibrium depend on parameter values.
Under a broad range of circumstances, however, positive effects of vegetation
density on both predator density and stability will be observed (compare equilibrium 1 with equilibria 2 and 3 in Fig. 2.7 A; see figure caption for details of the
model and the interpretation of isocline graphs).
Fig. 2.7 A assumes constant prey behavior across varying densities of predators
and vegetation. So far, only few studies have empirically addressed the long-term
effects of risk-sensitive behavior on the popUlation dynamics of macroinvertebrate
prey, as mediated through altered individual growth and fecundity (Peckarsky et
al., 1993; Ball and Baker, 1996). In theory, risk-sensitive prey behavior simultaneously bends the predator isocline to the right (prey hide more at higher
predator densities) and reduces prey production (prey feed less at higher predator
densities), both of which tend to stabilize the eqUilibrium (Ives and Dobson,
1987). Increasing vegetation density decreases the need for the prey to reduce its
own activity in the presence of the predator, because vegetation decreases the
abilities of fish predators to find prey. Because of its opposing effects on predator
efficiency and prey activity, the net effect of increasing vegetation density on the
equilibrium densities of predators and prey is not trivial to predict. Still, in models
of risk-sensitive foraging by the prey, the eqUilibria of predators and prey may not
depend on encounter rates (Abrams, 1984). Thus, vegetation may affect the
densities of predators and prey mostly through its positive effect on the productivity of the prey's resources. For a family of models, increasing the productivity
of the resources of a risk-sensitive prey increases the eqUilibrium densities of both
prey and predator (e.g., Abrams, 1984, Appendix A, B; Abrams, 1992, Appendix B). We conclude that, if macroinvertebrates are risk-sensitive foragers, this
will often increase the likelihood for both macroinvertebrates and fish to increase
with vegetation density.
Macroinvertebrates are a heterogeneous collection of taxa and size classes that
interact trophic ally with one another. Again, only few studies have, so far, empirically addressed the population dynamical consequences for macroinvertebrates of,
for example, omnivory and cannibalism (Diehl, 1993b, 1995; Hopper et al., 1996).
The accommodation of omnivory in the framework of Figure 2.7 requires, at
minimum, the addition of a third dimension to the phase space; that is, fish (P),
predatory (N), and nonpredatory (R) macroinvertebrates have to be represented
separately (Fig. 2.7C). For various reasons, the prediction of potential effects of
macrophytes on the dynamics of P, N, and R (as mediated through effects on
resource productivity and encounter rates) becomes very complex, and no simple
relationships can be derived for the (in nature extremely common) case of coexist-
2. Fish-Macroinvertebrate Interactions
37
ence of all three trophic levels. First, the stability conditions of omnivory systems
are complex, even in the simplest case of linear functional responses depicted in
Figure 2.7C, and the stability of the three-trophic level equilibrium can change
abruptly along smooth gradients of parameter changes (Holt and Polis, 1997).
Second, for stable three-trophic level equilibria, vegetation-induced increases of
the carrying capacity of R should increase both R and P but decrease N, and it
cannot be generally known whether total macroinvertebrates (R + N) will increase
or decrease with an increase in the carrying capacity of R (S. Diehl, unpublished
data). Finally, the effects of vegetation on equilibria mediated through changes in
encounter rates depend on how the benefit for N of reduced encounter rates with P
balances with potential changes in encounter rates with R. Cannibalism among
predatory macroinvertebrates can be qualitatively accommodated in the framework of Figure 2.7C by treating it as an extreme form of interference competition
(Wollkind, 1976). Thus, the N isocline would bend over to the right in the R plane
as cannibalism increases with the density of N. However, the prediction of possible effects of increasing vegetation density on equilibrium densities becomes
very complex, because vegetation now also affects the encounter rates among
conspecifics of N.
In conclusion, both scant empirical evidence and simple predator-prey models
suggest that increasing the density of submerged vegetation enhances the densities
of both macroinvertebrates and their fish predators. Assuming risk sensitivity in
the prey behavior tends to reinforce this conclusion, but the inclusion of increased
trophic complexity (omnivory, cannibalism) into models precludes the deduction
of simple, theoretical relationships among the abundances of macrophytes, macroinvertebrates, and fish. In the following section, we explore the implications of an
additional complexity (i.e., population size structure) for the interactions among
submerged macrophytes, macro invertebrates, and fish.
Vegetation, Macroinvertebrates, and Size Structure of Fish Populations
Most fish populations are size-structured, and many fish species feed on macroinvertebrates in littoral vegetation only during parts of their life histories (Werner,
1986; Persson, 1988). Unless all size classes of a fish species are affected similarly
by vegetation, the accommodation of size structure in the fish population goes
beyond the conceptual framework of Figure 2.7, because the density of macroinvertebrate-feeding fish would depend not only on the availability of macroinvertebrate resources but also on the availability of resources to all other life
stages of the fish (Mittelbach and Chesson, 1987; Persson, 1988; Osenberg et aI.,
1992; Mittelbach and Osenberg, 1993).
Many macroinvertebrate-feeding fish are vulnerable to piscivores and use
littoral vegetation as a predation refuge (Werner and Hall, 1988; EklOv and
Diehl, 1994; Persson and EklOv, 1995). Consequently, the presence of submerged macrophytes in lakes creates the opportunity for different size classes
of macroinvertebrate-feeding fish to segregate by habitat. In the presence of
piscivores, small vulnerable-size classes tend to seek refuge in the vegetation,
38
S. Diehl and R Kornij6w
i~
S
Body size
Figure 2.8. Schematic representation of the size specific individual growth rates of a fish
species that shifts from a vegetated to an open habitat at size S (solid line). The open habitat
is assumed to be more profitable than the vegetated habitat in terms of foraging return but
more dangerous in terms of risk of piscivory. Furthermore, the risk of mortality from
piscivores is assumed to decrease with increasing body size. At size S, the net benefit of
using the vegetation habitat becomes smaller than the net benefit of using the open habitat.
Also shown are the size-specific growth rates that would be observed if all size classes used
the open habitat in the absence of piscivores (broken line). The difference between the two
lines illustrates the instantaneous costs and benefits incurred by various size classes when
small fish use the vegetation habitat. Abundance and population size structure are assumed
to be identical in both cases (i.e., no feedbacks of differences in mortality and growth rates
on the population dynamics are taken into account).
where they often experience reduced individual growth rates while simultaneously
releasing larger invulnerable-size classes from intersize class competition in open
habitats (Werner et aI., 1983; Tonn et aI., 1992; Mittelbach and Osenberg, 1993;
Diehl and Ekl6v, 1995). These interactive effects of vegetation and piscivores on
the size-specific habitat use and growth rates of invertebrate-feeding fish suggest
a sharp increase in individual growth rate at the size of habitat switch (Fig. 2.8).
Such abrupt changes in growth rate have been documented for bluegill sunfish
(Osenberg et aI., 1988; Werner and Hall, 1988), and preliminary data indicate that
Eurasian perch may show similar discontinuities in their growth rates at the sizes
of ontogenetic niche shifts (L. Persson, personal communication).
The effects of submerged vegetation on patterns of habitat use and individual growth of fish are likely to feed back on the overall population
dynamics of the fish. For example, B lindow et al. (1991) observed positive
effects of increased submerged vegetation cover on both the growth rates of
macroinvertebrate-feeding perch and the abundance of piscivorous size classes
of perch. This example indicates the possibility that habitat segregation among
vulnerable and invulnerable size classes of fish in the presence of submerged
vegetation may counteract the stunting of fish populations. This has been documented for the bluegill sunfish, whose invulnerable stages attain high individual
growth rates in the open water when vulnerable stages are restricted to the
vegetation by piscivorous largemouth bass (Werner et aI., 1983; Mittelbach and
Osenberg, 1993). High individual growth rates of potentially piscivorous fish may,
in tum, feed back on the abundance of piscivores and thus on the relative risks of
vulnerable fish inside and outside the vegetation (Werner and Hall, 1988). However, the dynamics of size-structured fish populations may be rather complex even
2. Fish-Macroinvertebrate Interactions
39
in a single-habitat situation, and the population dynamical consequences of
changes in the relative sizes and qualities of the habitats used by different size
classes of fish are presently not well understood (Persson and Crowder, this
volume, Chapter 1).
Both vegetation and piscivores have to be present to produce habitat segregation
between vulnerable and invulnerable size classes of fish. However, the independent
occurrence of either abundant submerged vegetation or abundant piscivores seems to
be an uncommon and transient situation. Comparative data suggest, that the
abundance of piscivores in temperate European lakes is correlated with the abundance of submerged macrophytes (Grimm and Backx, 1990; Persson et aI., 1991;
Blindow et al., 1993; Persson, 1994). This relationship is likely to be causal, and
several positive feedback mechanisms between submerged vegetation and piscivores have been suggested (Blindow et al., 1993; Persson, 1994; Persson and Crowder, this volume, Chapter 1; Scheffer and Jeppesen, this volume, Chapter 31).
In conclusion, the question of which biomasses of macroinvertebrates and
macroinvertebrate-feeding fish are to be expected in a lake of a given productivity
and submerged macrophyte cover requires a complex answer. The relative and
absolute abundances of macroinvertebrate-feeding fish in open and vegetated
habitats depend on their habitat use as well as on the survival, growth, and
reproductive rates of all size classes of fish, which, in tum, depend on the
abundance of piscivores and on the relative sizes and productivities of open and
vegetated habitats (Mittelbach, 1984; Osenberg et aI., 1994; Persson et aI., 1997).
In a situation with both vegetation and piscivores, where predator-induced habitat
segregation among competing size classes of fish promotes slow growth of vulnerable
fish and fast growth of invulnerable fish, the densities of macroinvertebrate-feeding
fish and the resulting predation pressure on macroinvertebrates may be higher in
the vegetation and lower in open habitats than would be expected if each habitat
was the only one available. This tendency may be reinforced through population
dynamical feedback processes. For example, fast growth and high fecundity of
invulnerable size classes of fish, which are supported by resources outside the
vegetation, may result in far higher recruitment rates of small fish into the vegetation habitat than could be sustained by vegetation resources alone (Mittelbach and
Chesson, 1987; Mittelbach and Osenberg, 1993).
Suggestions for Future Research
The relationships among submerged macrophytes, macroinvertebrates, and fish
can be affected in complex ways by processes on a wide range of scales, from
behavioral decisions of individual macroinvertebrates to the life histories of longlived and wide-ranging piscivorous fish. StUdying the interactions of processes at
vastly differing scales is a challenging enterprise (Carpenter, 1988; Cooper et al.,
in press), and at present, the relative contributions of various processes to the
distributional patterns of macro invertebrates and fish are not well understood. To
date, empirical data are available predominantly on patterns at intermediate scales
40
s. Diehl and R Kornij6w
of spatial, temporal, and organismal resolution (e.g., within-season population
responses of macroinvertebrates to fish predation or diet and growth responses of
macroinvertebrate-feeding fish to piscivores and vegetation). Our knowledge
is especially deficient when it comes to the fine-scale mechanisms of fishmacroinvertebrate interactions and to the long-term population dynamics of
size-structured fish populations and of macroinvertebrate species with terrestrial
dispersal stages (e.g., insects).
In the future, high priority should be given to the study of behavioral interactions between macroinvertebrates and fish. This includes the systematic study of
the effects of submerged vegetation on the search behavior, prey choice, and
feeding and growth performances of different species and size classes of fish, as
well as the study of behavioral responses of macroinvertebrates to their invertebrate and fish predators (see, e.g., Mittelbach, 1981a; Huang and Sih, 1990). The
latter may provide crucial insights into the mechanisms of density dependence in
the individual growth of macroinvertebrate-feeding fish (see above). The study of
behavioral responses of macroinvertebrates to fish should be accompanied by the
investigation of the costs and benefits of these behaviors and their population
dynamical consequences to macroinvertebrate prey (Ball and Baker, 1996). In the
same vein, the behavioral trade-offs and population dynamical consequences of
predation and cannibalism among macroinvertebrates need to be studied (Peckarsky et al., 1993; Hopper et aI., 1996). The latter will require field experiments in
which the densities of selected macroinvertebrates are manipulated (Diehl, 1995).
Although many ofthe above processes are most profitably studied at laboratory
to field enclosure scales, long-term dynamics of fish and macroinvertebrates can
only be studied at the whole-system scale. Unfortunately, for the reasons listed in
the introduction, macroinvertebrates are most commonly not included in standard
lake surveys or the monitoring of lake manipulations. What is strongly needed are
both experimental and comparative studies of whole systems in which comprehensive data pertinent to the fish-macroinvertebrate interaction are collected. These
data should include estimates of (1) the proportional lake area covered with
macrophytes as well as macrophyte species composition and stand density; (2) the
species composition, abundance, size structure, and production of macroinvertebrates in habitats of varying vegetation cover and sediment structure; (3) the
species composition, abundance, size structure, and habitat distribution of fish as
well as their habitat- and size-specific diets and individual growth rates; (4) the
abundance and size structure of other important resources of fish, especially
zooplankton. Data should be gathered at several occasions during the year to allow
the study of seasonal patterns (Butler, 1989; Collins and Hinch, 1993). Also, to
increase statistical power and to facilitate comparisons among systems, it may
often be useful to lump macroinvertebrates into guilds of similar resource use or
vulnerability to fish predators (e.g., Diehl, 1992; Persson et al., 1996). Such sampling
programs may seem daunting, but the payoff in terms of the detection of significant
patterns can be high (see Blindow et al., 1993, for an excellent example).
Progress in the study of the long-term population dynamics of macroinvertebrates and fish can be significantly enhanced if empirical studies are comple-
2. Fish-Macroinvertebrate Interactions
41
mented by adequate modeling efforts. Because of the complexities of the life
histories of most macroinvertebrates and fish, simple modeling approaches (e.g.,
Fig. 2.7) can merely function as heuristic tools. The dynamics of prey with often
complex life cycles (e.g., insects) and predators that go through size-related
ontogenetic niche shifts need to be described by structured population models
(Crowley et aI., 1987; Mittelbach and Chesson, 1987; Persson et aI., 1997; Persson
and Crowder, this volume, Chapter 1). Although structured population models are
analytically far less tractable than simple strategic models, they hold promise to
increase the understanding of any particular real system under study by allowing
the incorporation of empirically derived (and testable) mechanistic detail (Murdoch et al., 1992). In this context, it is important to note that studies of fine-scale
mechanisms should be performed along gradients of conditions to guide the
choice of specific model functions (e.g., how does a specific parameter vary with
body size of fish and macroinvertebrates or with density of macrophytes?).
In this chapter, we have treated submerged vegetation largely as an independent
variable that sets the stage for the interactions between macroinvertebrates and
fish (Fig. 2.1). There are, however, various pathways of feedback through which
macroinvertebrates and fish affect submerged macrophytes (see Bronmark and
Weisner, 1992; Scheffer et aI., 1993; Lodge et aI., 1994; Bronmark and Vermaat,
this volume, Chapter 3; Jones et aI., this volume, Chapter 4; Lodge et aI., this
volume, Chapter 8; Scheffer and Jeppesen, this volume, Chapter 31). We have also
frequently simplified the macrophyte setting by contrasting two apparently
discrete scenarios, absence and presence of submerged macrophytes, without
specifying an exact level of the density or proportional cover of submerged
macrophytes for the latter situation. Our intention, here, was to illustrate clearly
distinct patterns and system behaviors. However, many of the processes discussed
in this chapter are likely to vary nonlinearly with the proportion of lake area
covered by submerged macrophytes (Scheffer et aI., 1993; Meijer et aI., 1994). An
especially important area of future research is therefore the exploration of patterns
and processes along gradients of vegetation density (e.g., Schriver et aI.,
1995). This bears the potential to identify critical levels of vegetation density
at which feedback mechanisms such as the ones dicussed in this and other
chapters of this book might become strong enough to stabilize lakes in states
favored by managers.
Acknowledgments. We thank the workshop organizers for making the meeting
very productive and enjoyable. We also thank Peter Abrams, Irmgard Blindow,
Scott Cooper, Lennart Persson, Marten Scheffer, and an anonymous reviewer for
comments on a previous draft of this chapter. Gary Mittelbach kindly provided the
original data to produce Figures 2.3A and 2.4A.
Financial support for the research on which parts of this chapter are based was
given by the Swedish Council for Forestry and Agricultural Research to S. Diehl
and L. Persson and by the British Council, the Finnish Academy of Sciences, the
Tor and Maj Nessling Foundation, and the International Agricultural Centre in
Wageningen to R. Kornij6w.
42
S. Diehl and R Kornij6w
References
Abrams, P.A. Foraging time optimization and interactions in food webs. Am. Nat. 124:8096; 1984.
Abrams, P.A. Predators that benefit prey and prey that harm predators: unusual effects of
interacting foraging adaptations. Am. Nat. 140:573-600; 1992.
Andersson, G.; Graneli, W.; Stenson, J. The influence of animals on phosphorus cycling in
lake ecosystems. Hydrobiologia 170:267-284; 1988.
Ball, S.L.; Baker, R.L. The non-lethal effects of predators and the influence of food
availability on life history of adult Chironomus ten tans (Diptera: Chironomidae). Freshwat. BioI. 34:1-12; 1995.
Ball, S.L.; Baker, R.L. Predator-induced life history changes: antipredator behavior costs or
facultative life history shifts? Ecology 77: 1116-1124; 1996.
Bendell, B.E.; McNicol, D.K. Fish predation, lake acidity and the composition of aquatic
insect assemblages. Hydrobiologia 150:193-202; 1987.
Bergman, E.; Greenberg, L.A. Competition between a planktivore, a benthivore, and a
species with ontogenetic niche shifts. Ecology 75: 1233-1245; 1994.
Blindow, I.; Andersson, G.; Hargeby, A. Submerged macrophytes in shallow lakes and their
importance for waterfowl. In: Finlayson, C.M.; Larsson, T., eds. Wetland management
and restoration. Proceedings of a workshop, 1990, Sweden. Solna, Sweden: Swedish
Environmental Protection Agency Report 3992; 1991:72-79.
Blindow, I.; Andersson, G.; Hargeby, A.; Johansson, S. Long-term pattern of alternative
stable states in two shallow eutrophic lakes. Freshwat. BioI. 30:159-167; 1993.
Bradford, D.F.; Cooper, S.D; Jenkins, T.M., Jr.; Brown, A.D.; Kratz, K.; Samelle, O.
Influences of natural acidity and introduced fish on amphibian, macroinvertebrate, and
zooplankton assemblages in alpine lakes of the Sierra Nevada, California. Can. J. Fish.
Aquat. Sci. (in press).
Bronmark, c.; Weisner, S.E.B. Indirect effects of fish community structure on submerged
vegetation in shallow, eutrophic lakes: an altemative mechanism. Hydrobiologia 2431244:
293-301; 1992.
Butler, M.J. Community responses to variable predation: field studies with sunfish and
freshwater macroinvertebrates. Ecol. Monog. 59:311-328; 1989.
Carpenter, S.R. Transmission of variance through lake food webs. In: Carpenter, S.R., ed.
Complex interactions in lake communities. New York: Springer-Verlag; 1988:119-135.
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Collins, N.C.; Hinch, S.G. Diel and seasonal variation in foraging activities of pumpkinseeds in an Ontario pond. Trans. Am. Fish. Soc. 122:357-365; 1993.
Cooper, S.D.; Diehl, S.; Kratz, K.; Samelle, O. Implications of scale for patterns and
processes in stream ecology. Aust. J. Ecol. (in press).
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegills and their prey. Ecology 63: 1802-1813; 1982.
Crowley, P.H.; Nisbet, R.M.; Guerney, W.S.c.; Lawton, J.H. Population regulation in
animals with complex life-histories: formulation and analysis of a damselfly model.
Adv. Ecol. Res. 17:1-59; 1987.
Cyr, H.; Downing, J.A. Empirical relationships of phytomacrofaunal abundance to plant
biomass and macrophyte bed characteristics. Can. J. Fish. Aquat. Sci. 45:976-984;
1988a.
Cyr, H.; Downing, J.A. The abundance of phytophilous invertebrates on different species of
submerged macrophytes. Freshwat. BioI. 20:365-374; 1988b.
Diehl, S. Foraging efficiency of three freshwater fish: effects of structural complexity and
light. Oikos 53:207-214; 1988.
Diehl, S. Fish predation and benthic community structure: the role of omnivory and habitat
complexity. Ecology 73:1646-1661; 1992.
2. Fish-Macroinvertebrate Interactions
43
Diehl, S. Effects of habitat structure on resource availability, diet and growth of benthivorous perch, Percafluviatilis. Oikos 67:403-414; 1993a.
Diehl, S. Relative consumer sizes and the strengths of direct and indirect interactions in
omnivorous feeding relationships. Oikos 68: 151-157; 1993b.
Diehl, S. Direct and indirect effects of omnivory in a littoral lake community. Ecology
76:1727-1740; 1995.
Diehl, S.; Eklov, P. Effects of piscivore-mediated habiat use on resources, diet, and growth
of perch. Ecology 76: 1712-1726; 1995.
Downing, J.A. Sampling the benthos dwelling on aquatic macrophytes. In: Downing, J.A.;
Rigler, F.H., eds. A manual of methods for the assessment of secondary productivity in
fresh waters. Oxford: Blackwell Scientific Publications; 1984:112-119.
EklOv, P.; Diehl, S. Piscivore efficiency and refuging prey: the importance of predator
search mode. Oecologia 98:344-353; 1994.
Erikson, M.O.G.; Henrikson, L.; Nilsson, B.-I.; Nyman, G.; Oscarson, H.G.; Stenson,
J.A.E. Predator-prey relations important for the biotic changes in acidified lakes. Ambio
9:248-249; 1980.
Evans, R.A. Response of limnetic insect populations of two acidic, fish less lakes to liming
and brook trout (Salvelinus fontinalis). Can. J. Fish. Aquat. Sci. 46:342-351; 1989.
Grimm, M.P.; Backx, 1.1.G.M. The restauration of shallow eutrophic lakes, and the role of
northern pike, aquatic vegetation and nutrient concentration. Hydrobiologia 200/201:
99-118; 1990.
Hill, A.M.; Lodge, D.M. Multi-trophic-Ievel impact of sublethal interactions between bass
and omnivorous crayfish. J. North Am. Benth. Soc. 14:306-314; 1995.
Holt, R.D.; Polis, G.A. A theoretical framework for intraguild predation. Am. Nat. 149:745764; 1997.
Hopper, K.R.; Crowley, P.H.; Kielman, D. Density-dependence, hatching synchrony, and
within-cohort cannibalism in young dragonfly larvae. Ecology 77:191-200; 1996.
Huang, c.; Sih, A. Experimental studies on behaviourally-mediated, indirect interactions
through a shared predator. Ecology 71 :1515-1522; 1990.
Ives, A.R.; Dobson, A.P. Antipredator behavior and the population dynamics of simple
predator-prey systems. Am. Nat. 130:431-447; 1987.
Jeffries, M. Invertebrate colonization of artificial pondweeds of differing fractal dimension.
Oikos 67:142-148; 1992.
Kajak, Z.; Kacprzak, K.; Polkowski, R. Tubular bottom sampler. Ekol. Pol. SeT. B II :159165; 1965.
Kornij6w, R. Cumulative consumption of the lake macrophyte Elodea by abundant generalist invertebrate herbivores. Hydrobiologia 319:185-190; 1996.
Kornij6w, R. The impact of predation by perch on the size-structure of Chironomus
larvae-the role of vertical distribution of the prey in the bottom sediments, and habitat
complexity. Hydrobiologia (in press).
Kornij6w, R.; Kairesalo, T. Elodea canadensis sustains rich environment for macroinvertebrates. Verh. Int. Verein. Limnol. 25 :4098-4111; 1994a.
Komij6w, R.; Kairesalo, T. A simple apparatus for sampling epiphytic communities associated with macrophytes. Hydrobiologia 294:141-143; 1994b.
Kornij6w, R.; Gulati, R.D.; van Donk, E. Hydrophyte-macroinvertebrate interactions
in Zwemhulst, a lake undergoing biomanipulation. Hydrobiologia 200/201:467474; 1990.
Kornij6w, R.; Gulati, R.D.; Ozimek, T. Food preference of some freshwater macroinvertebrates: comparing fresh and decomposed vascular plants and a filamentous alga.
Freshwat. BioI. 33:205-212; 1995.
Lillie, R.; Budd, J. Habitat architecture of Myriophyllum spicatum L. as an index to habitat
quality for fish and macroinvertebrates. J. Freshwat. Ecol. 7:113--125; 1992.
Lodge, D.M.; Barko, J.w.; Strayer, D.; Melack, J.M.; Mittelbach, G.G.; Howarth, R.W.;
Menge, B.; Titus, J.E. Spatial heterogeneity and habitat interactions in lake com-
44
S. Diehl and R Kornij6w
munities. In: Carpenter, S.R., ed. Complex interactions in lake communities. New York:
Springer-Verlag; 1988: 181-208.
Lodge, D.M.; Kershner, M.W.; Aloi, J.E.; Covich, A.P. Effects of omnivorous crayfish
(Oronectes rusticus) on a freshwater littoral food web. Ecology 75:1265-1281; 1994.
Macchiusi, F.; Baker, RL. Prey behaviour and size-selective predation by fish. Freshwat.
BioI. 25:533-538; 1991.
McPeek, M.A Determination of species composition in the Enallagma damselfly assemblages of permanent lakes. Ecology 71 :83-98; 1990a.
McPeek, M.A. Behavioural differences between Enallagma species (Odonata) influencing
differential vulnerability to predators. Ecology 71: 1714-1726; 1990b.
Meijer, M.-L.; Jeppesen, E.; van Donk, E.; Moss, B.; Scheffer, M.; Lammens, E.; van Nes,
E.; van Berkum, J.A; de Jong, GJ.; Faafeng, B.A; Jensen, J.P. Long-term responses to
fish stock reduction in small shallow lakes: interpretation of five-year results of four
biomanipulation cases in The Netherlands and Denmark. Hydrobiologia 2751276:457466; 1994.
Mittelbach, G.G. Foraging efficiency and body size: a study of optimal diet and habitat use
by bluegills. Ecology 62:1370-1386; 1981a.
Mittelbach, G.G. Patterns of invertebrate size and abundance in aquatic habitats. Can. J.
Fish. Aquat. Sci. 38:896-904; 1981b.
Mittelbach, G.G. Predation and resource partitioning in two sunfishes (Centrarchidae).
Ecology 65:499-513; 1984.
Mittelbach, G.G. Competition among refuging sunfishes and effects of fish density on
littoral zone invertebrates. Ecology 69:614-623; 1988.
Mittelbach, G.G.; Chesson, P.L. Predation risk: indirect effects on fish populations. In:
Kerfoot, W.e.; Sih, A., eds. Predation: direct and indirect impacts on aquatic communities. Hanover, NH: University Press of New England; 1987:315-332.
Mittelbach, G.G.; Osenberg, C.W. Stage-structured interactions in bluegill: consequences
of adult resource variation. Ecology 74:2381-2394; 1993.
Morin, PJ. The impact of fish exclusion on the abundance and species composition of larval
odonates: results of short-term experiments in a North Carolina farm pond. Ecology
65:53-60; 1984.
Murdoch, W.W; McCauley, E.; Nisbet, RM.; Guerney, W.S.e.; de Roos, AM. Individualbased models: combining testability and generality. In: DeAngelis, D.L.; Gross, L.J.,
eds. Individual-based models and approaches in ecology. London: Chapman & Hall;
1992:18-35.
Nelson, WG.; Bonsdorff, E. Fish predation and habitat complexity: are complexity thresholds
real? J. Exp. Mar. BioI. Ecol. 141:183-194; 1990.
Olson, M.H.; Mittelbach, G.G.; Osenberg, e.W. Competition between predator and prey:
resource-based mechanisms and implications for stage-structured dynamics. Ecology
76:1758-1771; 1995.
Osenberg, C.W.; Mittelbach, G.G. Effects of body size on the predator-prey interaction
between pumpkinseed sunfish and gastropods. Ecol. Monogr. 59:405-432; 1989.
Osenberg, C.W.; Mittelbach, G.G. The relative importance of resource limitation and
predator limitation in food chains. In: Polis, G.A; Winemiller, K.O., eds. Food webs:
integration of patterns and dynamics. New York: Chapman & Hall; 1996: 134-148.
Osenberg, e.W.; Werner, E.E.; Mittelbach, G.G.; Hall, D.J. Growth patterns in bluegill
(Lepomis macrochirus) and pumpkinseed (L. gibbosus) sunfish: environmental variation and the importance of ontogenetic niche shifts. Can. J. Fish. Aquat. Sci. 45: 17-26;
1988.
Osenberg, e.W; Mittelbach, G.G.; Wainwright, P.e. Two-stage life histories in fish: the
interaction between juvenile competition and adult performance. Ecology 73:255-267;
1992.
Osenberg, C.W.; Olson, M.H.; Mittelbach, G.G. Stage structure in fishes: resource productivity and competition gradients. In: Stouder, D.J.; Fresh, K.L.; Feller, RJ., eds. Theory
2. Fish-Macroinvertebrate Interactions
45
and application of fish feeding ecology. Columbia, SC: University of South Carolina
Press; 1994:151-170.
Pardue, WJ.; Webb, D.H. A comparison of aquatic macroinvertebrates occuring in association with eurasian watermilfoil (Myriophyllum spicatum) with those found in the open
littoral zone. J. Freshwat. Ecol. 3:69-79; 1985.
Peckarsky, B.L.; Cowan, C.A.; Penton, M.A.; Anderson, e. Sublethal consequences of
stream-dwelling predatory stoneflies on mayfly growth and fecundity. Ecology 74:
1836-1846; 1993.
Persson, L. Asymmetries in competitive and predatory interactions in fish populations. In:
Ebenman, B.; Persson, L., eds. Size-structured populations-ecology and evolution.
Heidelberg: Springer-Verlag; 1988:203-218.
Persson, L. Natural shifts in the structure of fish communities: mechanisms and constraints
on perturbation sustenance. In: Cowx, I.G., ed. Rehabilitation of freshwater fisheries.
Fishing News Books. Oxford: Blackwell Scientific Publishers; 1994:421-434.
Persson, L.; Eklov, P. Prey refuges affecting interactions between piscivorous perch and
juvenile perch and roach. Ecology 76:70-81; 1995.
Persson, L.; Greenberg, L.A. Juvenile competitive bottlenecks: the perch (Perea fluviatilis)roach (Rutilus rutilus) interaction. Ecology 71:44-56; 1990.
Persson, L.; Andersson, J.; Wahlstrom, E.; Eklov, P. Size-specific interactions in lake
systems: predator gape limitation and prey growth rate and mortality. Ecology 77:900911; 1996.
Persson, L.; Diehl, S.; Eklov, P.; Christensen, B. Size specific trade-offs and individual
behaviour-consequences at the population and community levels. In: Godin, J.-GJ.,
ed. Behavioural ecology of fishes. Oxford: Oxford University Press; 1997:316-343.
Persson, L.; Diehl, S.; Johansson, L.; Andersson, G.; Hamrin, S.F. Shifts in fish communities along the productivity gradient of temperate lakes-patterns and the importance of size-structured interactions. J. Fish BioI. 38:281-293; 1991.
Persson, L.; Diehl, S.; Johansson, L.; Andersson, G.; Hamrin, S.F. Trophic interactions in
lake ecosystems: a test of food chain theory. Am. Nat. 140:59-84; 1992.
Pieczynski, E. Numbers and biomass of the littoral fauna in Mikolajskie Lake and in some
other Masurian lakes. EkoI. Pol. 25:45-57; 1977.
Pierce, e.L.; Rasmussen, R.B.; Leggett, w.e. Littoral fish communities in southern Quebec
lakes: relationships with limnological and prey resource variables. Can. J. Fish. Aquat.
Sci. 51:1128-1138; 1994.
Polis, G.A.; Holt, R.D. Intraguild predation: the dynamics of complex trophic interactions.
Trends EcoI. EvoI. 7:151-154; 1992.
Prejs, A.; Koperski, P.; Prejs, K. Food web manipulation in small, eutrophic Lake Wirbel,
Poland: the effect of replacement of key predators on epiphytic fauna. Hydrobiologia
3421343:377-381; 1997.
Prejs, K. Bottom fauna. In: Pieczynski, E., ed. Selected problems of lake littoral ecology.
Warsaw: Warsaw University Press; 1976:23-144.
Rasmussen, J.B. Littoral zoobenthic biomass in lakes, and its relationships to physical,
chemical, and trophic factors. Can. J. Fish. Aquat. Sci. 45: 1438-1447; 1988.
Rasmussen, J.B. Patterns in the size structure of littoral zone macroinvertebrate communities. Can. J. Fish. Aquat. Sci. 50:2192-2207; 1993.
Savino, J.E; Marshall, E.A.; Stein, R.A. Bluegill growth as modified by plant density: an
exploration of underlying mechanisms. Oecologia 89:153-160; 1992.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends EcoI. EvoI. 8:275-279; 1993.
Schramm, H.L.; Jirka, K.J. Effect of aquatic macrophytes on benthic macroinverytebrates
in two Florida lakes. J. Freshwat. Eco!. 5: 1-12; 1989.
Schriver, P.; Brsgestrand, J.; Jeppesen, E.; Srsndergaard, M. Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large-scale enclosure experiments in a shallow eutrophic lake. Freshwat. BioI. 33:255-270; 1995.
46
S. Diehl and R Kornij6w
Suren, AM.; Lake, P.S. Edibility of fresh and decomposing macrophytes to three species of
freshwater invertebrate herbivores. Hydrobiologia 178: 165-178; 1989.
Tonn, W.M.; Paskowski, C.A; Holopainen, U. Piscivory and recruitment: mechanisms
structuring prey populations in small lakes. Ecology 73:951-958; 1992.
Turner, AM.; Mittelbach, G.G. Predator avoidance and community structure: interactions
among piscivores, planktivores, and plankton. Ecology 71:2241-2254; 1990.
Werner, E.E. Species interactions in freshwater fish communities. In: Diamond, J.; Case, T.,
eds. Community ecology. New York: Harper & Row; 1986:344-358.
Werner, E.E. Individual behavior and higher-order species interactions. Am. Nat. 140:S5S32; 1992.
Werner, E.E.; Gilliam, J.F.; Hall, D.J.; Mittelbach, G.G. An experimental test of the effects
of predation risk on habitat use in fish. Ecology 64: 1540--1548; 1983.
Werner, E.E.; Hall, D.J. Ontogenetic habitat shifts in bluegill: the foraging rate-predation
risk trade-off. Ecology 69:1352-1366; 1988.
Wiley, M.J.; Gorden, R.w.; Waite, S.w.; Powless, T. The relationship between aquatic
macrophytes and sport fish production in Illinois ponds: a simple model. North Am. J.
Fish. Manage. 3:111-119; 1984.
Wollkind, D.J. Exploitation in three trophic levels: an extension allowing intraspecific
carnivore interaction. Am. Nat. 110:431-447; 1976.
3. Complex Fish-Snail-Epiphyton
Interactions and Their Effects on
Submerged Freshwater Macrophytes
Christer Bronmark and Jan E. Vermaat
Introduction
Eutrophication of shallow freshwater and marine ecosystems has often resulted in
a drastic decline in the areal extension and biomass of submerged macrophytes
and a concomitant increase in the biomass of phytoplankton (e.g., Phillips et al.,
1978; Cambridge et aI., 1986; Hough et aI., 1989; Shepherd et al., 1989). Light
availability is usually the most important factor determining the distribution pattern, biomass, and production of submerged macrophytes (e.g., Chambers and
Kalff, 1985; Duarte et al., 1986), and it has been suggested that increasing
phytoplankton biomass due to higher nutrient input results in a reduction of
available light to a level at which net photosynthesis by submerged macrophytes is
impossible (e.g., Jupp and Spence, 1977; Jones et aI., 1983). However, Phillips et
al. (1978) suggested that macrophytes may disappear even when the bottom is
within the euphotic zone where light availability is adequate for photosynthesis.
Instead of shading by phytoplankton, they argued that increasing nutrient levels
stimulate epiphyton growth, which has a negative effect on the macrophyte host
through shading and competition for nutrients. Recent modifications of the model
of Phillips et al. (1978) also invoke epiphyton as a key factor in the decline of
submerged macrophytes (e.g., Silberstein et aI., 1986; Moss, 1989; Bronmark and
Weisner, 1992; Van Vierssen et al., 1994).
An increase in nutrient input during eutrophication does not necessarily result
in a gradual decline of submerged macrophytes (Moss and Leah, 1982). Instead,
47
48
C. Bronmark and J.E. Vermaat
observations (e.g., Timms and Moss, 1984; Balls et aI., 1989; Blindow et aI., 1992)
as well as theoretical models (e.g., Scheffer et aI., 1993) suggest the possibility of
alternate dominance of macrophytes or phytoplankton at comparable nutrient
loadings. These alternative states would need self-inforcing feedbacks to have
a reasonable degree of stability (Scheffer, 1989). Efficient periphyton removal
by grazers may be one of these feedback mechanisms: numerical and/or functional responses in the grazer guild may prevent a massive increase of
epiphytic growth. However, factors such as complex trophic interactions in
food chains involving molluscivores, snails, and epiphyton, changes in nutrient
turnover rates, and abiotic disturbances may affect the strength of the grazerepiphyton interaction and thus eventually affect the distribution of submerged
macrophytes.
In this chapter, we evaluate whether epiphyton removal by freshwater snails
can be a sufficiently powerful feedback mechanism. Several recent reviews cover
parts of the broader field of epiphyton interactions (e.g., Bronmark [1989] on
snail-epiphyton-macrophyte interactions and Stevenson et aI. [1996] on key
factors in benthic algal ecology). Here, we briefly dwell on snail-epiphytonmacrophyte interactions as a necessary basis for our main focus: complex indirect
interactions in realistic food chains including molluscivorous and piscivorous fish.
We consider the effect of biotic processes and nutrient availability as potential
feedback mechanisms affecting macrophyte growth and distribution.
Snail-Epiphyton Interactions
Epiphyton is grazed by an array of herbivorous invertebrates ranging in size from
minute epiphytic copepods to rather large pulmonate snails and crayfish (e.g.,
Cattaneo and Kalff, 1986; Botts, 1993). Snails are an important component of the
benthic community in many freshwater systems, and several experimental studies
have shown that snail grazing has strong effects on epiphyton biomass, species
composition, architecture, and productivity (see Bronmark, 1989, and Stevenson
et aI., 1996, for reviews). However, other taxa may also affect epiphyton through
herbivory (Table 3.1; see also Jones et aI., this volume, Chapter 4).
Different grazers may have different efficiencies in removing epiphyton, but
few comparative studies exist. Larger snails were effectively replaced by smaller
cladocerans, oligochaetes, and chironornids in exclosures with decreasing mesh
widths (Cattaneo and Kalff, 1986). Using p32, Kairesalo and Koskirnies (1987)
estimated grazing rates of snails, oligochaetes, and chironomids separately and
found that oligochaetes were the most important herbivore in terms of overall
grazing impact. In this case, the smaller oligochaetes were much more numerous
than the snails (Table 3.1). However, in other lakes snail densities are often higher
than the ones Kairesalo and Koskirnies (1987) found and may equal those of
oligochaetes (Soszka, 1975; Lalonde and Downing, 1992), and several papers
have suggested that grazers other than snails have a negligible impact on epiphytic
cover (Bronmark et aI., 1992; Underwood et aI., 1992; Daldorph and Thomas,
3. Fish-Snail-Epiphyton Interactions
49
Table 3.1. Epiphyton Grazing Rates and Community Grazing Impact by Snails and Other
Grazer Types a
Reference
Kairesalo & Koskimies (1987)
Grazer taxon
Lymnaea peregra
Oligochaetes
Chironomids
Jacoby (1985)
Vermaat (1994)
Theodoxus fluviatilis
Several snail species
Grazing rate
(mgAFDW
lind/day)
0.85
0.07
0.03
0.6
0.6 ± 0.1
Density
(n/m2)
Grazing impact
(mgAFDW
Im2/day)
25
213
17,150
4,300
450
400
866
93
270
240
aEpiphyton dry weight (DW) was converted to ash-free dry weight (AFDW) by using a conversion
factor of 0.5. Grazing rate from Vermaat (1994) is a pooled mean ± I SE over different species and
sizes (see Fig. 3.1); the snail density reported for Vermaat (1994) is the one estimated to be necessary
to suppress a epiphyton spring bloom in Lake Veluwe.
1995). Hence, at this time a simple generalization on the quantitative significance
of these different grazer guilds is not possible.
The seasonal timing of the presence of different grazers may differ substantially and consequently affect their impact on epiphyton. Chironomids often dominate
in early spring (Mason and Bryant, 1975; Cattaneo, 1983; Cattaneo and Kalff,
1986), but their impact is nullified after pupation. Snails, however, are present
year-round and may also remain active at low temperatures (Calow, 1975; Vermaat, 1994). The limited data available on snail community grazing rates suggest
that the overall grazing impact of snails is in the order of 200-300 mg ash-free dry
mass (AFDM)/m2/day (i.e., per unit periphyton covered surface), which should be
sufficient to suppress epiphytic spring blooms in eutrophic lakes (Van Vierssen et
aI., 1994). Ingestion rates in general are to a considerable extent governed by
consumer body size (e.g., Cammen, 1980), and this should also be expected for
freshwater snails. Cattaneo and Mousseau (1995) analyzed what factors affect
periphyton removal rates by invertebrate grazers and showed that grazer body
mass was most important. Similarly, Vermaat (1994) found that most of the
variation in periphyton removal rates between four snail species (Bithynia tentacuiata, Lymnaea peregra, Physa jontinaiis, and Valvata piscinalis) was explained by snail size (Fig. 3.1; ANOVA: P (species) = .762, P (covariable size) =
.037).
Intense grazing pressure may also reduce algal species diversity and affect the
three-dimensional structure of the epiphytic community (e.g., Steinman, 1996).
Typically, large overstory species such as stalked diatoms and filamentous algae
decline in response to grazing, whereas the proportion of small, tightly adhering
understory algae increases. Thus, grazing may halt or reverse the successional
process, keeping the algal community at an early seral stage of small prostate
algae.
C. Bronmark and J.E. Vermaat
50
:=I
"0
';"
3
y = O.093x-O.162
r2=O.1B. p=O.03
"0
.5
::!:
CI
IL
-<
T
2
•
1
CI
.§
•
-...
V
t il
a;
>
E
v
0
...
!
o
T
•1
r
0
5
.L
•
i:
•
•
T
!
10
15
snail shell height (mm)
Figure 3.1. Periphyton removal rate as a function of shell height of four common freshwater snail species: Bithynia tentaculata (L.), Lymnaea peregra (Miill.), Physa fontinalis
(L.) and Valvata piscinalis (Miill.). Shell lengths for these species were 8.3 ± 0.4,9.3 ± 0.8,
6.3 ± 0.8 and 4.0 ± 0.1, replication was 11, 9, 2, and 5, respectively. AFDM, ash-free dry
mass. (Data from Vermaat, 1994.)
Selective foraging may be a mechanism that could affect epiphyton species
composition when subjected to grazing snails. Several studies have shown that
snails show a preference for certain algal taxa (e.g., Calow, 1973, 1974), and it has
been suggested that specific algal preferences coupled to species-specific differences in activity of digestive enzymes are a way for freshwater snails to partition a
limiting resource between species (Calow and Calow, 1975). Kesler et al. (1986)
instead suggested that differences in the dentition of the snail rasping tongue, the
radula, are important in resource partitioning among sympatric snails. Bamese et
al. (1990) examined the grazing trails of pulmonate and prosobranch snails and
found no significant difference in grazing efficiency among pulmonate snails,
whereas the prosobranchs were less effective grazers. Limited differences in
radula morphology, as within pulmonate species, may thus not be important in
affecting grazing efficiency, whereas the larger differences found between pulmonates and prosobranchs may affect the impact of snails on periphyton algal
communities. However, the foraging apparatus of snails seems to operate at a
different spatial scale compared with individual algal cells (i.e., snails are not
able to discriminate and select specific components of the epiphyton community but, rather, are indiscriminate browsers). Given a patchiness in epiphyton composition, snails may choose to feed in patches or in microhabitats
with a high proportion of a preferred food item (Lodge, 1985, 1986; Vermaat,
1994).
3. Fish-Snail-Epiphyton Interactions
51
Snail grazing may also affect the productivity of epiphyton, and several studies
have shown that a decrease in epiphyte biomass due to grazing often is followed
by an increasing biomass-specific productivity (see review in Steinman, 1996).
Herbivores may enhance primary productivity by recycling limiting nutrients,
selecting for a shift in species composition toward faster-growing taxa, reduced
competition for light and nutrients for understory algae, and removal of dead and
senescent algal cells (e.g., Lamberti et aI., 1989; Steinman, 1996).
Epipbyton-Macropbyte Interactions
The significance of epiphyton-macrophyte interactions has received considerable
attention, especially the possibility of a symbiotic interaction between epiphytes
and their host plant (e.g., Wetzel, 1983, 1996; Burkholder, 1996). Epiphytic algae
have clear advantages of being associated with macrophytes. For example, epiphyton benefits from the macrophyte a priori by being provided with a large
surface area for colonization and an elevated position in the water column with a
higher availability of light. A more controversial issue has been the possibility of
exchange of different dissolved substances between epiphytic algae and macrophytes. Macrophytes have been found to leak: appreciable amounts of inorganic
nutrients and dissolved organic compounds (reviewed by Burkholder, 1996),
which should be readily used and beneficial to the algal and bacterial components
of the epiphytic community, especially in oligotrophic waters. The rate of leakage
changes with plant age, with an increasing release of high-quality substances as
the leaves start to senesce and decompose. The taxonomic divergence of epiphytic
communities on different species of macrophytes under low nutrient availability
has been suggested to be due to host -specific composition of the released organic
and inorganic compounds (Eminson and Moss, 1980). Thus, the epiphytic community clearly benefits by being associated with macrophytes, but it is less
obvious what the benefits of this association are for the macrophytes. Epiphytes
have been suggested to provide macrophytes with organic micronutrients (Wetzel,
1983), protection from pathogenic bacteria (Rogers and Breen, 1983), shading
under intense Mediterranean irradiance (Van Viers sen, 1983), or diversion of
grazers away from the macrophyte tissue (Hutchinson, 1975), but so far not much
experimental evidence exists to support these hypotheses.
On the contrary, increased epiphytic growth is generally associated with
negative impacts to the hosting macrophyte, often resulting in reduced growth.
Macrophyte photosynthesis and growth may be negatively affected by an
epiphytic cover through reduced light availability, an increased diffusive boundary layer hampering access to carbon and nutrient pools in the surrounding
water mass, detrimental oxygen and pH levels at the leaf surface, or anoxia
during darkness (e.g., Sand-Jensen, 1977; Bulthuis and Woelkerling, 1983;
Sand-Jensen and Borum, 1984; Silberstein et aI., 1986; Vermaat, 1994; Vermaat and Hootsmans, 1994).
52
C. Bronmark and J.E. Verrnaat
Snail-Epiphyton-Macrophyte Interactions
As described above, experimental studies have shown that snail grazing often
reduces the biomass of periphytic algae and that epiphytes may have an adverse
effect on the growth of their macrophyte host. Thus, here exists a potential for
complex indirect interactions between snails, epiphytes, and macrophytes. This
led J.D. Thomas and co-workers to suggest a close mutualistic relationship between freshwater snails and submerged macrophytes (e.g., Thomas, 1982; Thomas
et al., 1985; see also Carpenter and Lodge, 1986). It can therefore be hypothesized
that macrophytes should attract grazing snails that remove the epiphytic cover and
thereby benefit the macrophytes through decreasing competition for light and
nutrients. Grazing snails would not only benefit from the macrophyte association
by being provided with epiphytic food but may also be provided with a refuge
from predators, a substrate for oviposition, and an accessway to atmospheric
oxygen. Observational and experimental studies have verified several predictions
emanating from this hypothesis. Complex habitats provide a refuge from predators
by decreasing predation efficiency (Persson and Crowder, this volume, Chapter 1),
and snail communities in macrophyte beds have been shown to have higher
species richness as well as higher densities than surrounding nonvegetated littoral
areas (Fig. 3.2; Brbnmark, 1985a; Lodge, 1985; Lodge and Kelly, 1985; Lodge et
al., 1987; Brown and Lodge, 1993).
Brbnmark (l985b) showed in a laboratory experiment that the snail L. peregra
enhanced the growth rate of the submerged macrophyte Ceratophyllum demersum
by almost 30% compared with controls without snails. A small-scale field experi1000 ,--·······......·..·-..·· ....
•
---------------·r . . · ............·-·-......---·-..--..-............_-,
BltflymD
tentacu/ata
..s
/~·)O
B
POlarnc.pyr{]us
jenklnsi/
~
500
Va/veda
Oiscli7a/ls
250 '
a
1-\
OL-.... ------~~~_____
bare
submerse plants
reeds
habitat
Figure 3.2. Habitat associations in late winter in freshwater snails. Mean adult densities ±
SE are given. Significantly different habitat means are indicated with different lettering,
lower case for B. tentaculata, upper case for the other two species. Data were collected in
early 1994 from paired littoral transects with or without macrophytes in 13 lakes, ponds, and
canals in the western part of The Netherlands.
3. Fish-Snail-Epiphyton Interactions
53
ment by Underwood et al. (1992) provided similar results, whereas Underwood
(1991) suggested that increased turnover rates of nutrients by snails may explain
the increase in the macrophyte growth rate. Vermaat (1994) found no apparent
effect of snail grazing on growth enhancement of Potamogeton pectinatus, but a
more detailed analysis revealed that ungrazed plants with a thick epiphyton cover
produced more leaves by re-allocating resources from tubers. A high leaf production rate may compensate for reduced photosynthesis by older leaves heavily
burdened with epiphyton (cf. Sand-Jensen, 1977). Studies on marine seagrasses
(Hootsmans and Vermaat, 1985; Howard and Short, 1986; Williams and Ruckelshaus, 1993), and stream macroalgae (Dudley, 1992) have shown that epiphyton
grazing snails, insects, and crustaceans may enhance host growth in these environments as well.
Complex Food Chain Interactions
In recent years, several theoretical and empirical studies have emphasized the
importance of complex interactions for structuring freshwater communities (e.g.,
Kerfoot and Sib, 1987; Carpenter, 1988; Gulati et al., 1990). Most research on
indirect interactions in freshwater habitats has focused on interactions in pelagic
food chains. Experimental increase of piscivore density has often resulted in a
decrease in phytoplankton mediated through a decrease in the density of planktivorous fish and an increase in large efficient grazers (i.e., cladoceran zooplankton). Fewer studies on complex interactions have involved benthic food
chains, but lately several studies have demonstrated the existence of trophic
cascades and other complex interactions in these food chains as well (reviewed by
Bronmark et al., 1997). Below, we evaluate the importance of complex interactions for submerged macrophytes and start with the effects of predation on
freshwater snails, a key process in this respect.
Direct and Indirect Effects of Molluscivores on Snails
In a conceptual model, Lodge et ai. (1987) emphasized the importance of predation for structuring freshwater snail assemblages in permanent ponds and lakes
where abiotic constraints such as calcium concentration and/or disturbance events
(e.g., drying out) are not limiting. A range of predators prey on freshwater snails,
including invertebrates such as be10stomatid bugs, dytiscid larvae, leeches, flatworms, and crayfish and vertebrate predators such as fish and birds. Experimental
manipulations of predator density have shown that predators may reduce the
density of freshwater snails and also induce a shift in species composition toward
smaller, more hard-shelled species (Brown and DeVries, 1985; Weber and Lodge,
1990; Merrick et aI., 1991; Bronmark et aI., 1992; Martin et al., 1992; Osenberg et
aI., 1992; Bronmark 1994). These studies have documented a direct lethal effect of
predation, but freshwater snails may also be able to evaluate local predation risk
and respond behaviorally. In laboratory studies, predatory crayfish have been
C. Bronmark and lE. Vermaat
54
PS
No P$
50
40
;;R
~
(j)
30
:3
(j)
0)
.2
20
rf
10
0
Cont rol
FIsh cue
Snail cue
Figure 3.3. (A) The proportion of snails (all species) occupying the covered part of
artificial substrates placed in lakes with (PS) or without pumpkinseed sunfish (no PS).
Vertical bars denote 1 SE. (Modified from Turner, 1996.) (B) Proportion of Physella snails
using refuges when exposed to water with chemical cues from pumpkinseed sunfish (fish
cues), to water in which snails had been crushed (snail cue), and to water without cues
(control). Vertical bars denote 1 SE. (Modified from Turner, 1996.)
shown to induce pulmonate snails to crawl out of the water to decrease encounter
rates with the predator (Alexander and Covich, 1991), and further, snails respond
to some leeches by crawl-outs, shell-shaking, closing their operculum, or decreasing their activity (Townsend and McCarthy, 1980; Bronmark and Malmqvist,
1986). Changes in activity patterns may be state-dependent, however, as in B.
tentaculata, which showed a higher reduction of activity in response to leeches
when food was absent compared with when periphyton was present in the experimental tanks (Bronmark and Malmqvist, 1986). Turner (1996) compared snail
3. Fish-Snail-Epiphyton Interactions
55
habitat use in lakes with and without pumpkinseed sunfish (Lepomis gibbosus), a
specialized snail predator, and found that snails spent more time under cover in
lakes with pumpkinseeds (Fig. 3.3A). In a pool experiment, he was able to confmn
that presence of cover increased snail survival, and further, a laboratory experiment showed that the snail Physella sp. changed its behavior toward increased
refuge use in response to chemical cues from crushed conspecifics (Fig. 3.3B).
Fish cue did not in itself elicit a change in behavior. McColltHn et al. (unpublished
data; see also Crowder et al., this volume, Chapter 14) also found that chemical
cues from a snail-eating fish resulted in reduced activity levels of snails. Besides
changes in behavior, presence of predators may also affect the life history
strategies of prey. Crowl and Covich (1990) found that waterborne chemical cues
from crayfish feeding on snails induced a change in snail size at first reproduction
and also affected their longevity. Snails exposed to cues from predatory crayfish
started to reproduce later and at a larger size than snails that were not exposed to
predatory crayfish. Presence of molluscivores may thus cause phenotypic changes
in life history strategies in snails, affecting their population density and structure.
Food Chain Effects of Molluscivores
Several recent studies have tested whether changes in the density of molluscivorous predators will cascade down and eventually affect the biomass of primary
producers. Bronmark et al. (1992) manipulated the density of a highly specialized
molluscivorous fish, the pumpkinseed sunfish, in enclosures in two North American lakes. The density and biomass of snails and periphyton were monitored over
two seasons and compared with fish exclosures and cageless controls. Pumpkinseed sunfish dramatically reduced the density and biomass of snails, which in tum
resulted in an increased biomass of epiphyton due to the reduced grazing pressure.
Martin et al. (1992) reported similar results from a study designed to test the
effects of large and small sunfish (especially Lepomis micr%phus, redear sunfish)
on littoral macroinvertebrates. They found strong negative effects of fish on the
biomass of snails and a positive effect on periphytic algae. The effect was independent of sunfish size, which was counter to the original predictions. The
authors had predicted that small sunfish would only affect smaller macroinvertebrates such as soft-bodied insect larvae and that only larger sunfish were expected to be able to decimate snail populations (cf. Mittelbach, 1984; Stein et al.,
1984). The strong effect from even small sunfish was suggested to be due to
predation on newly hatched snails early in the season, preventing the recruitment
of snails to larger sizes, and thus, a predator may have long-term effects on lower
trophic levels through a short-term predation event by influencing the recruitment
of a dominant herbivore (snails). What is even more interesting in this context is
that Martin et al. (1992) found an effect of sunfish on submerged macrophytes as
well. In the second year, submerged macrophytes (Najas and Potamogeton)
sprouted in cages without fish, resulting in a dramatic difference in comparison
with cages with fish and with the lake bottom outside the cages where hardly any
macrophytes were found. The change in macrophyte abundance was attributed
56
C. Bronmark and J.E. Vermaat
to reduced negative effects of epiphyton when snails became abundant in the
cages without molluscivorous fish. Bronmark (1994) found analogous results
in a enclosure/exclosure experiment involving a cyprinid fish, the tench (Tinea
tinea). Tench has the morphological capacity to crush snail shells with its molariform pharyngeal teeth, but previous studies had suggested that tench diets only
include a minor proportion of snails (e.g., Kennedy and Fitzmaurice, 1970),
indicating that tench effects on lower trophic levels would be weak. The field
experiment clearly showed, however, that tench may have strong effects on interactions in benthic food chains. Snails were dramatically reduced in tench enclosures compared with either cages with perch (Percafluviatilis) or cages with no
fish, and the resulting decrease in grazing pressure allowed for an increase in the
biomass of epiphytic algae (Fig. 3.4). Further, growth of the submerged macrophyte Elodea canadensis was significantly reduced in tench cages, possibly due to
increased competition for light and nutrients with the epiphytic algae.
Crayfish is another important and common snail predator that may affect
interactions in benthic food chains involving snails and epiphyton. However,
crayfish are more omnivorous, feeding on macroinvertebrates, macrophytes, periphyton, and detritus. Omnivory should decrease interaction strength in interaction
chains, as direct effects may be counteracted by indirect effects (e.g., Strong,
1992). Given the omnivorous diet, it is not intuitively obvious how manipulations
in crayfish density would affect lower trophic levels. A reduction of snails in the
presence of crayfish may be predicted to have a positive effect on the biomass of
epiphytic algae, but crayfish may, however, have a direct negative effect on
epiphyton through grazing. Further, the effect of crayfish on submerged macrophytes may be through indirect effects on epiphyton or by direct consumption.
Recent! y, Lodge et al. (1994) tested the strength of direct and indirect interactions
imposed by crayfish in littoral benthic food chains. Crayfish were found to have a
strong negative effect on snails and submerged macrophytes, whereas there was a
positive effect on epiphyton biomass per unit macrophyte surface area, indicating
that crayfish grazing effects were not as strong as the indirect snail-epiphyton
effect. Although an increased epiphyton biomass would be expected to have a
negative effect on the growth of submerged macrophytes, a high occurrence of
floating fragments of submerged macrophytes suggested that direct consumption
explained most of the decline of macrophytes in crayfish enclosures.
The above studies have shown strong cascading effects in benthic food chains
based on a direct lethal effect of predators on herbivores. However, as seen above,
predators may also affect the behavior of prey, and a change in activity or habitat
use in snails in response to presence of molluscivores may be expected to affect
snail-epiphyton interaction strength. Turner (1977) manipulated the perceived
risk of predation by adding different amounts of crushed snails to experimental
pools. The perceived increase in risk of predation resulted in an increased use of
cover by the snail Physella, and this had positive effects on the biomass of
periphytic algae in habitats outside the cover, whereas periphyton biomass in
refuge was kept at low levels (Fig. 3.5). Increased risk of predation also resulted in
a reduced growth of Physella.
57
3. Fish-Snail-Epiphyton Interactions
f
N
E
"~
0
0
10
8
OJ
........
Cf)
Cf)
T
SNAILS
6
nI
E
0
4
.Cij
2
co
c:
Vl
f
~~
I
I
t
0
........
N
500
E
u 400
PERIPHYTON
.-
"~
0
u. 300
«
OJ
2>
c:
200
....
0
~
c.
.t:
T
100
Q)
0..
0
........
150
N
E
MACROPHYTES
"~
o
OJ
........
100
Cf)
Cf)
nI
E
0
co
50
C1I
~
.s:!
IJ.j
0
Control
Tench
Figure 3.4. Direct and indirect effects of molluscivorous tench (Tinea tinea) on snails,
periphyton, and the submerged macrophyte Elodea canadensis. (Modified from Bronmark,
1994.)
c. Bronmark and J.E. Vermaat
58
70
-r--
........
~
........
A
-.--
Q>
(fI
:::>
+'"
IV
+'"
:a
IV
60
T
:r:
c:
Q>
-.--
a.
0
50
o
0.25
Perceived
N'3
~
4
predation
risk
B
~
Cl
LL
«
2
OJ
E
........
~
IV
E
o
i:D
c:
o
>,
..c:
a.
.;::
~
0
o
0.25
Perceived
4
predation
risk
Figure 3.5. (A) Snail habitat use and (B) periphyton biomass in open and covered habitats
in relation to perceived mortality risk. Predation risk was manipulated by adding different
amounts of crushed snails to experimental pools (zero to four snails per day).
Piscivore Effects Down Benthic Food Chains
Bronmark and Weisner (1992) suggested that the change between alternative
stable states in shallow eutrophic lakes was due to stochastic abiotic disturbance
events acting on fish community structure. Inherent in this model was the importance of strong direct links between piscivorous and molluscivorous fish, which
then cascaded down the food chain and affected macrophytes through changes in
3. Fish-Snail-Epiphyton Interactions
59
abundance of herbivorous snails and epiphyton. Very few studies have evaluated
the effect of piscivores down benthic food chains with four trophic levels. The
only experimental study of piscivore effects in such food chains was performed by
Power (1990), who found strong cascade effects in a stream system with a food
chain involving piscivorous fish, benthivorous fish, grazing invertebrates, and
periphytic algae. Piscivores reduced the abundance of predatory fish fry and
invertebrates, which resulted in an increase of a herbivorous chironomid and a
decrease of algae. Bronmark and Weisner (1996) took another approach. Instead
of experimental manipulations, they surveyed a large number of natural ponds for
patterns in the distribution of piscivorous and molluscivorous fish, snails, and
periphytic algae. The ponds differed only with respect to fish assemblage structure
and were divided into three categories: ponds without fish, ponds with molluscivorous fish, and ponds with both molluscivorous and piscivorous fish. Ponds with
two trophic levels (no fish) had a high density and biomass of snails and a low
biomass of periphyton, whereas ponds with three trophic levels (with molluscivorous fish) had reduced densities and biomasses of snails and a high biomass of
periphyton. This is in accordance with predictions from food chain theory and the
earlier findings from experimental manipulations of molluscivore density (see
above). However, ponds with four trophic levels (with piscivores) deviated from
the predicted pattern. Although densities of molluscivorous fish were low and
densities of snails were high, the biomass of periphyton was high and did not differ
from ponds with only molluscivorous fish. A closer examination of the size
structure of molluscivorous fish in ponds with and without piscivores suggested a
possible explanation. Ponds with piscivores were dominated by sparse populations
of large-bodied molluscivorous tench and crucian carp, whereas in ponds without
piscivores, the tench and crucian carp populations were very dense and dominated
by small individuals (Bronmark et al., 1995). Piscivores are gape-limited predators, and thus there exists an upper prey size above which prey no longer are
vulnerable to predators (i.e., prey have reached an absolute size refuge). Selective
predation by large molluscivores that had reached an absolute size refuge from
piscivory reduced the density of large snail species (lyrnneids) that are effective
grazers, resulting in a snail community dominated by small detritivorous snails.
Thus, decoupling at the piscivore-molluscivore level resulted in no cascading
effects ofpiscivores on periphytic algae in this system (see Hambright et aI., 1991;
Hambright, 1994, for examples of decoupling due to size refuges in pelagic food
chains), and consequently, changes in the abundance of piscivores may have no, or
negligible, effects on the biomass of submerged macrophytes through indirect
interactions in the benthic food chain.
Nutrient Effects
Nutrient availability is an important determinant of epiphyton biomass, species
composition, and productivity. In some systems, nutrient availability is the primary determinant of algal biomass (bottom-up control), whereas in others grazing
60
C. Bronmark and J.E. Vermaat
is the major factor controlling algae (top-down control). Other factors such as light
availability and disturbance events may be of great importance in yet other
systems (see Stevenson et aI., 1996, for a thorough review of factors affecting
freshwater benthic algae). Here, we first consider how biotic processes may
interact with nutrient availability by modifying turnover rates and then look at a
larger scale and see how changes in the productivity of a system interact with food
chain interactions and what the consequences might be.
Nutrient Turnover
Recent studies have shown that consumers, besides having direct lethal effects,
may affect primary producers through a change in the availability and quality of
nutrients (reviewed by Vanni, 1996). Consumers make nutrients that were immobilized in prey biomass available to primary producers by excretion or defecation.
However, consumers may not only affect the quantity of available nutrients but
may also through their excretions change the relative availability or the ratio of
nutrients, and this may affect algal species differently (Vanni, 1996). Several
studies have experimentally shown that grazing by snails and other benthic macroinvertebrates may affect periphyton through changes in nutrient turnover rates.
Mulholland et al. (1983) showed that grazing on epiphyton increased nutrient
turnover rates, and Underwood (1991) found that presence of snails enhanced the
growth of the submerged macrophyte Ceratophyllum even if they were not in
contact with the plant. The release of phosphate, nitrate, nitrite ammonia, and urea
by snails was suggested to increase phosphorus and nitrogen levels, resulting in an
increase in plant growth. Underwood (1991) estimated the release rates to be 3.7 x
10-4 jlg phosphate, 23.9 x 10-4 jlg nitrate, 4.4 x 10-4 jlg ammonia, and 0.8 x 10-4 jlg
urea per hour and milligrams total wet weight of snail. For comparison, at a
density of 100 individuals/m2, adult Planorbis planorbis (Underwood, 1991; Underwood et aI., 1992) would then release up to 24 jlmol phosphate/m2/day, which
is in the same order of magnitude as phosphate release from oxic sediments
(10 jlmollm2/day) (Nlirnberg, 1984). Cuker (1983), however, found no increase in
nutrient availability by grazing snails and suggested that nutrients may be recycled
in the grazer gut system by algal cells resistant to gut passage. However, the
significance of nutrient release from grazer feces must depend on grazer density,
grazer nutrient conservation possibilities, and the nutrient content of the food.
Therefore, in oligotrophic arctic lakes (Cuker, 1983; Merrick et aI., 1991) with low
snail densities, fecal nutrient release must be low and difficult to quantify as
compared with more eutrophic systems (e.g., Underwood, 1991; Underwood et
al.,1992).
A laboratory experiment has also suggested that nutrient recycling by fish may
affect periphyton biomass and species composition (McCollum and Crowder,
unpublished data), and fish-mediated changes in phytoplankton have been
ascribed to changes in nutrient recycling and availability (e.g., Threlkeld, 1987).
Observations by Bronmark et al. (1992) and Bronmark (1994) may, however,
suggest that this mechanism can be refuted. Comparisons of periphyton biomass
3. Fish-Snail-Epiphyton Interactions
61
between cages with pumpkinseed sunfish and yellow perch (Bronmark et al.,
1992) or tench and perch (Bronmark, 1994) showed a strong cascade effect
(fish-snails algae) in pumpkinseed and tench cages, whereas the biomass of
periphyton in perch cages was identical to cages without fish. Thus, increased
nutrient turnover rates due to fish excretion did not affect periphyton biomass.
Environmental Productivity
The overall productivity of an environment is one of the most important factors
determining the abiotic template that biotic processes can operate within. Theoretical models have suggested that potential length of food chains in an environment is set by environmental productivity (e.g., Oksanen et al., 1981; Fretwell,
1987). The Oksanen model also predicts that effects of increases in environmental
productivity on equilibrium biomass will depend on the total number of trophic
levels in the environment and if the population under consideration is a primary
producer, herbivore, or primary or secondary carnivore (i.e., the trophic level).
Several studies in the laboratory and in the field have shown that enrichment
with phosphorus and/or nitrogen results in increased biomass of benthic algae
(reviewed by Borchardt, 1996). For example, phosphorus enrichment of small
areas in the littoral zone of an oligotrophic North American lake resulted in a
dramatic increase in epiphytic biomass (Osenberg, 1989; see also Wetzel, 1996).
Growth of snails increased in response to the nutrient enhancement, and it was
concluded that both snails and epiphytes were resource limited (Osenberg, 1989).
Other studies have shown a smaller effect of nutrient enhancement on algal
biomass. Instead, the increase in nutrients seems to have been channeled into
increased grazer production, resulting in grazers maintaining constant algal biomasses despite nutrient enhancement (e.g., Hill et aI., 1992). The observed differences in response to nutrient enhancement may be explained by at least two
mechanisms. First, there is a problem with the temporal scale of experimental
manipulations. Algae and their grazers operate under very different temporal
scales, with algae responding numerically to changes in resource levels in the
course of days to weeks, whereas the reproductive cycle of snails and other
herbivores is more in the order of months up to years and thus sets limits to the
numerical responses that could be observed in short-term nutrient enrichment
experiments. The importance of temporal scale was clearly shown in an experiment in which a pristine tundra river was fertilized for four consecutive summers
(Peterson et aI., 1993). A dramatic increase of benthic algal biomass and productivity was found during the first two summers, but this response was modified in
the last years due to an increase of herbivorous insects that prevented a buildup of
periphytic algae. Further, the number of trophic levels in the food chain may affect
the response of the primary producers to nutrient enhancement. In food chains
with an even number of trophic levels, primary producers are controlled by
herbivores, and addition of nutrients should be transferred to an increased herbivore biomass, whereas plant biomass should remain constant (e.g., Oksanen et
aI., 1981; Fretwell, 1987). In food chains with only one level, grazers obviously
62
C. Bronmark and I.E. Vermaat
cannot control plant biomass, and in three-level systems predators keep herbivores
at such a low level that they do not control plants. Thus, in odd-numbered food
chains nutrient enhancement is predicted to result in increasing plant biomass. No
such manipulation has been performed with benthic systems in lakes, but Rosemond et al. (1993) found that nutrient enhancement in a stream system resulted in
increased algal biomass in treatments without grazers (one trophic level) and no
effect on algal biomass in treatments with grazers (two trophic levels). Similarly,
Wootton and Power (1993) manipulated primary productivity in a stream system,
not by nutrient enhancement but through changes in light availability, and found
that algae responded positively to increased productivity in food chains with three
trophic levels, but when a fourth trophic level (piscivores) was added grazers
increased and algae decreased. Thus, temporal scale and food chain configuration
may both affect the response to nutrient enhancement in freshwater systems.
Conclusions
Many conceptual models have been proposed to explain the changes in submerged
macrophyte distribution in shallow eutrophic lakes and in several complex inter-
,------------
i
___ .::!~~e______ ~t
I
I
~ I
I
~------------
oI
~
: i
E I
~
PISCIVORES
I
I
SMALL
LARGE
·r ~
MOLLUSCIVORES ............ MOLLUSCIVORES
:__ !e!u~e____
I
, - - - - - : - - - - - - - - - SNAILS
i
I
i
I
MACROPHYTES
r
EPIPHYTON
i -><1
' - - - ..... NUTRIENTS
LIGHT
Figure 3.6. Complex interactions in a benthic freshwater food chain. Filled lines denote
direct consumer-resource interactions. Hatched lines emanating in macrophytes represent
an interaction modification (i.e., macrophytes add increased habitat complexity, which
modifies the strength of the consumer-resource interaction by decreasing predator efficiency). Hatched lines from consumers denote nutrient recycling (e.g., excretions, defecation),
whereas punctuated line show recruitment of small molluscivorous fish into the larger size
classes where they have reached an absolute size refuge from predation by gape-limited
piscivores.
3. Fish-Snail-Epiphyton Interactions
63
actions between grazers, epiphyton, and plants are central components. As reviewed above, several empirical studies have shown that many of the direct
interactions between these actors in benthic freshwater food chains are strong and
may even give rise to strong indirect interactions as predicted by food chain
theory. However, these predictions seem to hold only for systems with up to three
trophic levels. This is problematic because most lakes hold populations of piscivorous fish. Experiments on pelagic food chains have suggested that changes in
piscivore densities may cascade all the way down to primary producers, but it
seems that absolute size refuge in the primary carnivore and changes in the species
composition of the dominant herbivores may decouple strong cascading interactions from piscivores in benthic food chains (Bronmark and Weisner, 1996).
Further, feedback loops such as nutrient recycling by herbivores and predators and
the provision of prey refuge by structurally complex macrophyte beds (Fig. 3.6)
may further decrease our ability to understand or predict the outcome of changes
in benthic food chains. Thus, despite considerable progress in understanding food
chain interactions in benthic freshwater systems, our possibilities to predict food
chain dynamics in complex littorals are still rather limited. Future experimental
studies including simultaneous manipulations of resource and consumer levels, as
well as experiments performed at larger temporal and spatial scales, are of key
importance.
Acknowledgments. We thank Francien Boessenkool for access to unpublished data
(Fig. 3.2); Larry Crowder, Beth McCollum, and Andy Turner for access to unpublished manuscripts; and Dave Lodge and J.D. Thomas for comments on an
earlier manuscript. And, of course, the organizers of the workshop! Financial
support was received from the Swedish Council for Forestry and Agricultural
Research (CB).
References
Alexander, J.E., Jr; Covich, AP. Predator avoidance by the freshwater snail Physella
virgata in response to the crayfish Procambarus simulans. Oecologia 87:435-442;
1991.
Balls, H.; Moss, B.; Irvine, K. The loss of submerged plants with eutrophication. I. Experimental
design, water chemistry, aquatic plant and phytoplankton biomass in experiments carried out
in ponds in the Norfolk Broadlands. Freshwat. BioI. 22:71-87; 1989.
Barnese, L.E.; Lowe, RL.; Hunter, RD. Comparative grazing efficiency of pulmonate and
prosobranch snails. J. North Am. Benth. Soc. 9:35-44; 1990.
Blindow, I.; Andersson, G.; Hargeby, A; Johansson, S. Long-term patterns of alternative
stable states in two shallow eutrophic lakes. Freshwat. BioI. 30:159-167; 1992.
Borchardt, M.A Nutrients. In: Stevenson, RJ.; Bothwell, M.L.; Lowe, RL., eds. Algal
ecology. Freshwater benthic systems. San Diego, CA: Academic Press; 1996; 183-227.
Botts, P. The impact of small chironomid grazers on epiphytic algal abundance and dispersion. Freshwat. BioI. 30:25-33; 1993.
Bronmark, C. Freshwater snail diversity: effects of pond area, habitat heterogeneity and
isolation. Oecologia67:127-131; 1985a.
Bronmark, C. Interactions between epiphytes, macrophytes and herbivores: an experimental approach. Oikos 45:26-30; 1985b.
64
e. Bronmark and J.E. Vermaat
Bronmark, C. Interactions between epiphytes, macrophytes and freshwater snails: a review.
J. Moll. Stud. 55:299-311; 1989.
Bronmark, C. Effects of tench and perch on interactions in a freshwater, benthic food chain.
Ecology 75:1818-1824; 1994.
Bronmark, e.; Malmqvist, B. Interactions between the leech Glossiphonia eomplanata and
its gastropod prey. Oecologia 69:268-276; 1986.
Bronmark, e.; Weisner, S.E.B. Indirect effects of fish community structure on submerged
vegetation in shallow, eutrophic lakes: an alternative mechanism. Hydrobiologia 243/
244:293-301; 1992.
Bronmark, e.; Weisner, S.E.B. Decoupling of cascading trophic interactions in a freshwater, benthic food chain. Oecologia 108:534--541; 1996.
Bronmark, e.; Paszkowski, e.A.; Tonn, W.M.; Hargeby, A Predation as a determinant of
size structure in populations of crucian carp (Carassius earassius) and tench (TInea
tinea). Eco!. Freshwat. Fish. 4:85-92; 1995.
Bronmark, C.; Klosiewski, S.P., Stein, R.A Indirect effects of predation in a freshwater,
benthic food chain. Ecology 73: 1662-1674; 1992.
Bronmark, e.; Dahl, J.; Greenberg, L.A Complex trophic interactions in freshwater benthic
food chains. In: Streit, B.; SHidler, T.; Lively, C., eds. Ecology and evolution offreshwater animals. Basel: Birkhauser Publishers; 1997:55-88.
Brown, K.M.; DeVries, D.R. Predation and the distribution and abundance of a pulmonate
pond snai!. Oecologia 66:93-99; 1985.
Brown, K.M.; Lodge, D.L. Gastropod abundance in vegetated habitats: the importance of
specifying null models. Limno!. Oceano gr. 38:217-225; 1993.
Bulthuis, D.A; Woelkerling, w.J. Biomass accumulation and shading effects of epiphytes
on leaves of the seagrass, Heterozostera tasmanica, in Victoria, Australia. Aquat. Bot.
16:137-148; 1983.
Burkholder, J.M. Interactions of benthic algae with their substrata. In: Stevenson, R.J.;
Bothwell, M.L.; Lowe, R.L., eds. Algal ecology. Freshwater benthic systems. San
Diego, CA: Academic Press; 1996:253-297.
Calow, P. The food of Ancylus fluviatilis (Miill), a littoral, stone-dwelling herbivore.
Oecologia 13:113-133; 1973.
Calow, P. Evidence for bacterial feeding in Planorbis contortus Linn. (Gastropoda: Pulmonata). Proc. Malac. Soc. Lond. 41 :145-156; 1974.
Calow, P. The feeding strategies of two freshwater gastropods, Ancylus fluviatilis (Miill.)
and Planorbis contortus Linn. (Pulmonata), in terms of ingestion rates and absorption
efficiencies. Oecologia 20:33-49; 1975.
Calow, P.; Calow, L.J. Cellulase activity and niche separation in freshwater gastropods.
Nature 255:478-480; 1975.
Cambridge, M.e.; Chiffings, A.W.; Brittan, e.; Moore, L.; McCombe, AJ. The loss of
seagrass in Cockburn Sound, Western Australia. II. Possible causes of seagrass decline.
Aquat. Bot. 24:269-285; 1986.
Cammen, L.M. Ingestion rate: an empirical model for aquatic deposit feeders and detritivores. Oecologia 44:303-310; 1980.
Carpenter, S.R., ed. Complex interactions in lake communities. New York: Springer-Verlag;
1988.
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Cattaneo, A Grazing on epiphytes. Limno!. Oceanogr. 28:124--132; 1983.
Cattaneo, A; Kalff, J. The effect of grazer size manipulation on periphyton communities.
Oecologia 69:612-617; 1986.
Cattaneo, A; Mousseau, B. Empirical analysis of the removal rate ofperiphyton by grazers.
Oecologia 103:249-254; 1995.
Chambers, P.A; Kalff, J. Depth distribution and biomass of submersed macrophyte communities in relation to Secchi depth. Can. J. Fish. Aquat. Sci. 42:701-709; 1985.
3. Fish-Snail-Epiphyton Interactions
65
Crowl, T.A; Covich, AP. Predator-induced life-history shifts in a freshwater snail. Science
247:949-951; 1990.
Cuker, B.E. Grazing and nutrient interactions in controlling the activity and composition of
the epilithic algal community of an arctic lake. Limnol. Oceanogr. 28:133-141; 1983.
Daldorph, P.W.G.; Thomas, J.D. Factors influencing the stability of nutrient-enriched
freshwater macrophyte communities: the role of sticklebacks Pungitius pungitius and
freshwater snails. Freshwat. BioI. 33:271-289; 1995.
Duarte, C.M.; Kalff, J.; Peters, R.H. Patterns in biomass and cover of aquatic macrophytes.
Can. J. Fish. Aquat. Sci. 43:1900-1908; 1986.
Dudley, T.L. Beneficial effects of herbivores on stream macroalgae via epiphyte removal.
Oikos 65:121-127; 1992.
Eminson, D.; Moss, B. The composition and ecology of periphyton communities in freshwaters. 1. The influence of host type and external environment on community composition. Br. Phycol. J. 15:429-446; 1980.
Fretwell, S.D. Food chain dynamics: the central theory of ecology? Oikos 50:291-301;
1987.
Gulati, R.D.; Lammens, E.H.R.R.; Meijer, M.-L.; van Donk, E., eds. Biomanipulation. Tool
for water management. Dordrecht: Kluwer Academic Publishers; 1990.
Hambright, K.D. Morphological constraints in the piscivore-planktivore interaction: Implications for the trophic cascade hypothesis. Limnol. Oceanogr. 39:897-912; 1994.
Hambright, K.D.; Drenner, R.W.; McComas, S.R.; Hairston, N.G., Jr. Gape-limited piscivores, planktivore size refuges, and the trophic cascade hypothesis. Hydrobiologia
121:389-404; 1991.
Hill, W.R.; Boston, H.L.; Steinman, AD. Grazers and nutrients simultaneously limit lotic
primary productivity. Can. 1. Fish. Aquat. Sci. 49:504-512; 1992.
Hootsmans, M.J.M., Vermaat, J.E. The effects of periphyton-grazing by three epifaunal
species on the growth of Zostera marina L. under experimental conditions Aquat. Bot.
22:83-88; 1985.
Hough, R.A.; Fomwall, M.D.; Negele, B.J.; Thompson, R.L.; Putt, D.A Plant community
dynamics in a chain of lakes: principal factors in the decline of rooted macrophytes with
eutrophication. Hydrobiologia 173:199-217; 1989.
Howard, R.K.; Short, ET. Seagrass growth and survivorship under the influence of epiphyte
grazers. Aquat. Bot. 24:287-302; 1986.
Hutchinson, G.E. A treatise on limnology. III. Lirnnological botany. New York: Wiley;
1975.
Jacoby, J.M. Grazing effects on periphyton by Theodoxus fluviatilis (Gastropoda) in a
lowland stream. J. Freshwat. Ecol. 3:265-274; 1985.
Jones, R.e.; Walti, K.; Adams, M.S. Phytoplankton as a factor in the decline of the
submersed macrophyte Myriophyllum spicatum L. in Lake Wingra, Wisconsin, USA
Hydrobiologia 107:213-219;1983.
Jupp, B.; Spence, D.H.W. Limitations on macrophytes in a eutrophic lake, Loch Leven. I.
Effects of phytoplankton. 1. Ecol. 65:175-186; 1977.
Kairesalo, T.; Koskimies, I. Grazing by oligochaetes and snails on epiphytes. Freshwat.
BioI. 17:317-324; 1987.
Kennedy, M.; Fitzmaurice, P. The biology of Tench, Tinea tinea (L.) in Irish waters. Proc.
R. Irish Acad. 69:31-82; 1970.
Kerfoot, W.e.; Sih, A, eds. Predation. Direct and indirect impacts on aquatic communities.
Hanover, NH: University Press of New England; 1987.
Kesler, D.H.; Jokinen, E.H.; Munns, W.R., Jr. Trophic preferences and feeding morphology
of two pulmonate snails species from a small New England pond, USA Can. J. Zool.
64:2570-2575; 1986.
Lalonde, S.; Downing, J.A Phytofauna of eleven macrophyte beds of differing trophic
status, depth and composition. Can. J. Fish. Aquat. Sci. 49:992-1000; 1992.
66
C. Bronmark and J.E. Vermaat
Lamberti, G.A.; Ashkenas, L.R; Gregory, S.V.; Steinman, A.D.; McIntire, C.D. Productive
capacity of periphyton as a determinant of plant-herbivore interactions in streams.
Ecology 70:1840-1856; 1989.
Lodge, D.M. Macrophyte gastropod associations: observations and experiments on macrophyte choice by gastropods. Freshwat. BioI. 15:695-708; 1985.
Lodge, D.M. Selective grazing on periphyton: a determinant of freshwater gastropod
microdistribution. Freshwat. BioI. 16:831-841; 1986.
Lodge, D.M.; Kelly, P. Habitat disturbance and the stability of freshwater gastropod populations. Oecologia 68:111-117; 1985.
Lodge, D.M.; Brown, K.M.; Klosiewski, S.P.; Stein, R.A.; Covich, A.P.; Leathers, B.K.;
Bronmark, C. Distribution of freshwater snails: spatial scale and the relative importance
of physicochemical and biotic factors. Am. Maiacoi. Bull. 5:73-84; 1987.
Lodge, D.M.; Kershner, M.W.; Aloi, J. Effects of an omnivorous crayfish (Orconectes
rusticus) on a freshwater littoral food web. Ecology 75: 1265-1281; 1994.
Martin, T.H.; Crowder, L.B.; Dumas, c.P.; Burkholder, J.M. Indirect effects of fish on
macrophytes in Bays Mountain Lake: evidence for a littoral trophic cascade. Oecologia
89:476-481; 1992.
Mason, e.P.; Bryant, R.I. Changes in the ecology of the Norfolk Broads. Freshwat. BioI.
5:257-270; 1975.
Merrick, G.W.; Hershey, A.E.; McDonald, M.E. Lake trout (Salvelinus namaycush) control
of snail density and size distribution in an arctic lake. Can. J. Fish. Aquat. Sci. 48:498502; 1991.
Mittelbach, G.G. Predation and resource partitioning in two sunfishes (Centrarchidae).
Ecology 65:499-513; 1984.
Moss, B. Water pollution and the management of ecosystems: a case study of science and
scientists. In: Grubb, P.I.; Whittaker, J.B., eds. Toward a more exact ecology. Oxford:
Blackwell; 1989:401-422.
Moss, B.; Leah, R.T. Changes in the ecosystem of a guanotrophic and brackish shallow lake
in Eastern England: potential problems in its restoration. Int. Rev. Ges. Hydrobiol.
67:625-659; 1982.
Mulholland, P.I.; Newbold, J.D.; Elwood, J.w.; Hom, C.L. The effect of grazing intensity
on phosphorus spiralling in autotrophic streams. Oecologia 58:358-366; 1983.
Nilrnberg, G. The prediction of internal phosphorus load in lakes with anoxic hypolimnion.
Limnol. Oceanogr. 29:111-124; 1984.
Oksanen, L.; Fretwell, S.D.; Arruda, J.; Niemela, P. Exploitation ecosystems in gradients of
primary productivity. Am. Nat. 118:240-261; 1981.
Osenberg, C.w. Resource limitation, competition and the influence of life history in a
freshwater snail community. Oecologia 79:512-519; 1989.
Osenberg, C.w.; Mittelbach, G.G.; Wainwright, P.c. Two-stage life histories in fish: the
interaction between juvenile competition and adult performance. Ecology 73:255-267;
1992.
Peterson, B.J.; Deegan, L.; Helfrich, J.; Hobbie, J.E.; Hullar, M.; Moller, B.; Ford, T.E.;
Hershey, A.; Hiltner, A.; Kipphut, G.; Lock, M.; Fiebig, D.M.; McKinley, V.; Miller,
M.e.; Vestal, J.R; Ventullo, R.; Volk, G. Biological responses of a tundra river to
fertilization. Ecology 74:653-672; 1993.
Phillips, G.L.; Eminson, D.E; Moss, B. A mechanism to account for macrophyte decline in
progressively eutrophicated waters. Aquat. Bot. 4: 103-125; 1978.
Power, M.E. Effects offish in riverfood webs. Science 250:811-814; 1990.
Rogers, K.H.; Breen, e.M. An investigation of macrophyte, epiphyte and grazer interactions. In: Wetzel, RG., ed. Periphyton of freshwater ecosystems. The Hague: Junk
Publishers; 1983:217-226.
Rosemond, A.D., Mulholland, P.J.; Elwood, J.W. Top-down and bottom-up control of
stream periphyton: effects of nutrients and herbivores. Ecology 74: 1264-1280; 1993.
Sand Jensen, K. Effect of epiphytes on eelgrass photosynthesis. Aquat. Bot. 3:55-63; 1977.
3. Fish-Snail-Epiphyton Interactions
67
Sand Jensen, K.; Borum, J. Epiphyte shading and its effect on photosynthesis and diel
metabolism of Lobelia dortmanna L. during the spring bloom in a Danish lake. Aquat.
Bot. 20:109-119; 1984.
Scheffer, M. Alternative stable states in eutrophic, shallow freshwater systems: a minimal
model. Hydrobiol. Bull. 23:73-83; 1989.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evol. 8:275-279; 1993.
Shepherd, S.A.; McComb, A.J.; Bulthuis, D.A.; Neverauskas, V.; Steffensen, D.A.; West, R.
Decline of seagrasses. In: Larkum, A.W.D.; McComb, A.J.; Shepherd, S.A., eds. Biology of seagrasses, a treatise on the biology of seagrasses with special reference to the
Australian region. Aquatic Plant Studies. Vol. 2. Amsterdam: Elsevier; 1989:346-393.
Silberstein, K.; Chiffings, A.W.; McComb, A.I. The loss of seagrass in Cockburn Sound,
Western Australia. III. The effect of epiphytes on the productivity of Posidonia australis
Hook. F. Aquat. Bot. 24:355-371; 1986.
Soszka, G.J. The invertebrates on submerged macrcophytes in three Masurian lakes. Ekol.
Pol. 23:371-391; 1975.
Stein, R.A.; Goodman, C.G.; Marshall, E.A. Using time and energetic measures of cost in
estimating prey value for fish predators. Ecology 65:702-715; 1984.
Steinman, A.D. Effects of grazers on freshwater benthic algae. In: Stevenson, R.J.; Bothwell, M.L.; Lowe, R.L., eds. Algal ecology. Freshwater benthic systems. San Diego,
CA: Academic Press; 1996:341-373.
Stevenson, R.I.; Bothwell, M.L.; Lowe, R.L., eds. Algal ecology. Freshwater benthic
systems. San Diego, CA: Academic Press; 1996.
Strong, D.R. Are trophic cascades all wet? Differentiation and donor-control in speciose
ecosystems. Ecology 73:747-754; 1992.
Thomas, J.D. Chemical ecology of the snail hosts of schistosomiasis: snail-snail and
snail-plant interactions. Malacologia 22:81-91; 1982.
Thomas, J.D.; Nwanko, D.I.; Sterry, P.R. The feeding strategies of juvenile and adult
Biomphalaria glabrata (Say) under simulated natural conditions and their relevance to
ecological theory and snail control. Proc. R. Soc. Lond. B 226: 177-209; 1985.
Threlkeld, S. T. Experimental evaluation of trophic-cascade and nutrient mediated effects of
planktivorous fish on plankton community structure. In: Kerfoot, C.w.; Sih, A., eds.
Predation: direct and indirect impacts on aquatic communities. Hanover, NH: University
Press of New England; 1987:161-173.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Townsend, C.R.; McCarthy, T.K. On the defence strategy of Physa fontinalis (L.), a
freshwater pulmonate snail. Oecologia 46:75-79; 1980.
Turner, A.M. Freshwater snails alter habitat use in response to predation. Anim. Behav.
51:747-756; 1996.
Turner, A.M. Contrasting short-term and long-term effects of predation risk on consumer
habitat use and resource dynamics. Behav. Ecol. 8:120--125;1997.
Underwood, G.J.C. Growth enhancement of the macrophyte Ceratophyllum demersum in
the presence of the snail Planorbis planorbis: the effects of grazing and chemical
conditioning. Freshwat. BioI. 26:325-334; 1991.
Underwood, G.J.C.; Thomas, J.D.; Baker, J.H. An experimental investigation of interactions in snail-macrophyte-epiphyte systems. Oecologia 91 :587-595; 1992.
Vanni, M.J. Nutrient transport and recycling by consumers in lake food webs: implications
for algal communities. In: Polis, G.A.; Winemiller, K.O., eds. Food webs. Integration of
patterns and dynamics. New York: Chapman & Hall; 1996:81-95.
Van Viers sen, W. The influence of human activities on the functioning of macrophytedominated aquatic ecosystems in the coastal area of Western Europe. Proc. EWRS/AAB
Int. Symp. Aquat. Macrophytes Nijmegen; 1983:273-281.
68
C. Bronmark and J.E. Vermaat
Van Vierssen, W.; Hootsmans, M.J.M.; Vermaat, J.E., eds. Lake Veluwe, a macrophytedominated system under eutrophication stress. Dordrecht: Kluwer Academic Publishers;
1994.
Vermaat, J.E. Periphyton removal by freshwater micrograzers. In: van Vierssen, W.; Hootsmans, M.1.M.; Vermaat, J.E., eds. Lake Veluwe, a macrophyte-dominated system under
eutrophication stress. Dordrecht: Kluwer Academic Publishers; 1994:213-249.
Vermaat, J.E.; Hootsmans, M.J.M. Periphyton dynamics in a temperature-light gradient. In:
Van Vierssen, W.; Hootsmans, M.J.M.; Vermaat, J.E., eds. Lake Veluwe, a macrophytedominated system under eutrophication stress. Dordrecht: Kluwer Academic Publishers;
1994: 193-212.
Weber, L.M.; Lodge, D.M. Periphytic food and predatory craytish: relative roles in determining snail distribution. Oecologia 82:33-39; 1990.
Wetzel, R.G. Attached algal-substrata interactions: fact or myth, and when and how? In:
Wetzel, R.G., ed. Periphyton of freshwater ecosystems. The Hague: Dr. W. Junk Publishers; 1983:208-215.
Wetzel, R.G. Benthic algae and nutrient cycling in lentic freshwater systems. In: Stevenson,
R.1.; Bothwell, M.L.; Lowe, R.L., eds. Algal ecology. Freshwater benthic systems. San
Diego, CA: Academic Press; 1996:641-667.
Williams, S.L.; Ruckelshaus, M.H. Effects of nitrogen availability on eelgrass (Zostera
marina) and epiphytes. Ecology 74:904-918; 1993.
Wootton, J.T.; Power, M.E. Productivity, consumers, and the structure of a river food chain.
Proc. Natl. Acad. Sci. U.S.A. 90:1384-1387; 1993.
4. Interactions between Periphyton,
Nonmolluscan Invertebrates, and
Fish in Standing Freshwaters
John I. Jones, Brian Moss, and Johnstone O. Young
Introduction
The Importance of Periphyton
Densities of several hundreds of micrograms of algal chlorophyll a per square
meter are not unusual on submerged plant surfaces. Frequently, the supporting
surface cannot be seen through the covering mass, which may compete with the
plant for light, inorganic carbon, and nutrients. Given the problems posed by the
overlying water column as well as the periphyton (syn: epiphyton) for such plants,
it is remarkable that submerged plants develop at all in other than the clearest,
most nutrient-deficient waters.
That they do so abundantly, even in nutrient-rich waters, may therefore depend
on balancing mechanisms that prevent the development of both periphyton and
phytoplankton to the potentials set by the nutrient supply. Phytoplankton growth
can be lessened by grazing, allelopathy, washout, and mixing to depths where
respiration dominates photosynthesis. Given a substratum on which to develop,
only the former two mechanisms are likely to be important controls on the
periphyton, and to date there has been no unequivocal evidence in support of
allelopathy (Forsberg et aI., 1990). It is with the first, grazing, that this chapter is
concerned.
Macrophytes greatly increase the surface area available for colonization by
animals when compared with the lake bed beneath (see Diehl and Komij6w, this
69
70
J.1. Jones, B. Moss, and J.~. Young
volume, Chapter 2). Furthermore, periphyton provides more nutritious substrates
than sediments or vascular plant tissues. Consequently, periphyton is an important
energy source for both detritus and grazing food chains (Gressens, 1995). However, grazer links are likely to be complex for they involve a system of several
metabolically active components, including the plants themselves, the periphyton,
invertebrate and vertebrate grazers of a wide range of sizes and types, and the
invertebrate and vertebrate predators of these grazers. This review is concerned
with the control potentially exerted by such a system.
Organization and Scope of This Review
The review is organized in sections that take components of the periphyton grazer
system in turn: (1) the nature of the periphyton and (2) of the grazers; (3) grazer
effects on the periphyton; (4) periphyton effects on the grazers; (5) effects of
grazers on grazers; and (6) effects of predators on grazers. Finally, some general
observations are made.
Most studies on grazer-periphyton relationships have been in streams (Feminella and Hawkins, 1995) with less information from still waterbodies, where
emphasis has been on snail grazing. The impact of smaller invertebrates such as
microcrustacea, oligochaetes, and chironomid larvae on periphyton is not so well
known, although their high numbers and production in the littoral zone (Prejs,
1976; Sarvala et al., 1981) suggest a high potential grazing pressure (Cattaneo,
1983; Kairesal0, 1984; Kairesalo and Koskimies, 1987). This review excludes
work on snails, which are considered by Bronmark and Vermaat (this volume,
Chapter 3). References to work on streams are indicated by asterisks. Aloi (1990)
has recently reviewed the methodology used in investigations of periphyton.
Nature of the Periphyton
Communities colonizing new substrates often consist of low posture, prostrate, or
apically attached cells. With time, communities become progressively more threedimensional as protrusive, stalked, and/or filamentous forms grow (Patterson and
Wright, 1986*; Steinmann and McIntire, 1986*, 1987*). The resultant periphytic
communities are usually several layered, with an adherent understory of basal
cells of eventually filamentous algae, encrusting green algae, and tightly attached
diatoms such as Cocconeis. Overtopping it are protrusive long diatoms, stalked
diatoms, and short filaments. There may be further longer, loosely associated
filaments and colonies, and large unicells, sometimes motile, entangled in the
matrix. Different growth forms will have different susceptibilities to grazing. The
overstory is usually most vulnerable to grazers (Kessler, 1981; Jacoby, 1985*,
1987*; Steinmann et aI., 1987*).
Clouds of filamentous green algae (e.g., of Cladophora, Ulothrix, or Zygnematales) can form by proliferation of filaments formerly attached to substrates.
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
71
Although more precisely termed metaphyton, they often develop from and are
closely linked with epiphytic communities and are considered here also.
Nature of the Grazers
General Survey
Invertebrate grazers other than molluscs, in still waterbodies, include oligochaetes
(e.g., Young, 1945; Kairesalo and Koskimies, 1987; Hann, 1991), nematodes
(Kairesalo, 1984), microcrustacea (Cladocera, Copepoda, Ostracoda [e.g., Young,
1945; Fairchild et al., 1989; Gressens, 1995]), amphipod crustaceans (e.g., Mazumder et aI., 1989; Dodds, 1991), isopod crustaceans (e.g., Sozska, 1975), crayfish
(Flint and Goldman, 1975), mysids (Irvine et aI., 1993), chironomid larvae (e.g.,
Young, 1945; Mason and Bryant, 1975; Hann, 1991), caddisfly larvae (e.g.,
Sozska, 1975), and mayfly larvae (e.g., Moss, 1976; Sozska, 1975).
The impact of other littoral invertebrates grazers such as protozoans, haliplid
coleopterans, and corixid hemipterans on periphyton communities has not been
studied. McCormick (1991 *) and Sleigh et al. (1992*) have studied the first of
these groups in streams and found annual production to be of the same order as
macroinvertebrates and fish. Only a few papers on grazing on periphyton by fish
in still waterbodies are available (Cattaneo, 1983; Cattaneo and Kalff, 1986).
Amphibians at the larval stage may be major grazers (Osborne and McLachlan,
1985; Seale, 1980; Bronmark et al., 1991).
Most grazers are mobile, focusing their grazing on distinct patches of periphyton (e.g., Kohler, 1984*; Vaughn, 1986*; Scimgeoar et aI., 1991 *). A few are
sedentary, including some chironomid, caddisfly, and lepidopteran species, which
live in constructed retreats or shallow mines in macrophytes, and feed in a
localized area (Lamberti and Moore, 1984*; Hart, 1985*; Bergey, 1995*). Studies
of these species have been limited to lotic habitats.
Mouthpart Adaptations and the Nature of
the Food Eaten
The relative success or failure of herbivores in cropping periphyton is linked with
the suitability of their mouthparts to the physiognomy of the algal assemblage
(Fig. 4.1). Most grazers are scrapers (e.g., snails, which have a radula, and
chironomids, caddisflies, and mayflies, which have blade-like mandibles and
other mouthparts with limited setation). Some grazers are collector-gatherers
(some species of mayfly larvae), with mouthparts covered with dense brushes
of hairs or setae, most associated with the maxillae and labium, which allow
them to browse on periphyton. Other taxa use additional limbs to remove
periphyton from the substrate before feeding on it. In stream studies, differences in mouthpart morphology have been used to explain different effects on
periphyton (e.g., Lamberti et aI., 1987a*,b*; Karouna and Fuller, 1992*).
72
J.1. Jones, B. Moss, and J.~. Young
Figure 4.1. Scanning electron micrographs showing the scale and diversity of the mouthparts and associated feeding structures of six potential grazers. (Top left) Blade-like mouthparts of chironomid larvae (Cricotopus sylvestris) and (top right) caddis larvae (Limnephilidae);
(middle left) reduced beak-like mouthparts and front limbs of corixid hemipteran (Sigara
dorsalis); (middle right) setation of mouthparts of mayfly larvae (Cloeon dipterum); (bottom left) additional thoracic limbs, which assist in feeding of amphipod (Gammarus pulex),
and (bottom right) my sid (Neomysis integer). gp, gnathopod; 1, labrum; la, labium; m,
mentum; mn, mandible; mp, mandibular palp; mx, maxilla, pp, pereiopod; tl, thorax limbs.
Scale bars are as shown.
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
73
Grazer Effects on the Periphyton
Biomass
Grazer effects may include direct consumption, alteration of community composition, nutrient mobilization, physical disruption of the mat, and an increase in
growth rate and turnover of periphyton. Some stream studies have indicated that
certain algae typical of the early stages of colonization can benefit from grazers
under certain conditions (e.g., Lamberti and Resh, 1983*; Jacoby, 1987*; Petersen,
1987*) or that grazing may benefit the algal community as a whole and not just
individual species (e.g., in increased primary productivity [Lamberti and Resh, 1983*]
and species richness [Eichenberger and Schlatter, 1978*; Sumner and McIntire,
1982*]). However, much evidence shows that grazers can effectively remove
periphyton, and it has been suggested that a summer minimum in periphyton
biomass is brought about by grazers (Cattaneo and Kalff, 1986; Cattaneo, 1990).
One of the problems that exists in the interpretation of grazing studies is that
they often infer grazer effects from the difference in biomass between grazed and
ungrazed treatments. Due to the exponential nature of biomass accumulation, this
may not be true and especially so, if other density-dependent factors are limiting
the ungrazed biomass (Mitchell and Wass, 1996; Mitchell and Perrow, this volume, Chapter 9). This may explain findings such as those of Knudson (1957), who
concluded that grazing had little effect on density of an epiphytic diatom, Tabellaria flocculosa, growing on emergent plants in three English Lake District lakes,
as long as growth conditions were favorable; periods of constant density, however,
may represent a balance when the rate of increase equals the erosion rate (a
combination of grazing and wave action). Gressens (1995) also found that grazers
(chironomids, snails, chydorid cladocerans) did not significantly decrease biomass
of periphyton attached to artificial substrates. Hence, Gressens and Lowe (1994)
advise caution in the interpretation of correlations of chironomid density with low
periphyton biomass as they are not necessarily caused by grazing.
An accurate estimate of grazing rate is difficult to determine. Nevertheless,
Kairesalo and Koskimies (1987) found that the daily consumption by oligochaetes
and snails corresponded to 22-45% of the average phosphorus uptake by epiphytes on Equisetum fluviatile and concluded that grazing is an important
mechanism affecting epiphytic abundance. Botts and Cowell (1992) found that a
small chironomid, Psectocladius, ate 13.9% of the available Cosmarium standing
stock daily. Cattaneo (1983) used estimates of grazer standing stock (mainly,
oligochaetes and chironomids) and changes in epiphyte biomass to calculate
grazing rates sufficient to be a significant component of epiphyte decline. Data of
Mason and Bryant (1975), reworked by Kessler (1981), suggest grazing rates of
chironomids on Typha stems to have been 44% of the periphyton net accumulation
rate. All these studies indicate that invertebrate grazers can play an extremely
important role in controlling periphyton populations in lakes.
However, some reported cases do not fit with this general pattern. Fairchild et
al. (1 989a) found a positive relationship between invertebrate densities (especially
74
J.I. Jones, B. Moss, and J.O. Young
10,000 [
>-
ro
'0
1,000
L-
a>
N
ro
Cl
Cl
,•
E
I
10\
•
• •
••
ro...
ro>
E
a>
a:::
o1
001 I
0.01
•
• •
•• I•
,•
.
: •
••
•
() 1
.f'
••
•
-,•
•
•
••
I
......
a>
0
I
•
100
L-
• I,I ••
•
• •• ••
•
I
•
10
100
Grazer body mass (mg)
1,000
10.000
Figure 4.2. Relationship between estimated removal rate (see text for details) of periphyton and the grazer body mass. (From Cattaneo and Mousseau, 1995.)
chironomid larvae and chydorid dadocerans) and periphyton standing crop. They
speculate that the moderate algal consumption by grazers in a softwater lake with
low dissolved inorganic carbon (DIC) is accompanied by increased growth stimulated by regenerated carbon. Under similar conditions, Fairchild et al. (1989b)
found that removal of grazers did not increase periphyton density on porous pots
from which nutrients were diffusing.
While ignoring the problems associated with estimates of grazing rate, Cattaneo and Mousseau (1995) in a review of the literature found that uncorrected
values of periphyton removal rate increased linearly with grazer body mass and
food availability and decreased with grazer abundance (Fig. 4.2):
log R = 0.99log(bodymass) - 0.71 log (grazer biomass)
+ 0.46log(periphyton biomass)? = 0.78, P < .0001.
(1)
In order of importance, the correlates were grazer body mass (65% of variation),
crowding (7%), and food availability (6%). High grazer densities were associated
with competition for a limited food resource and increased aggressive behavior.
Temperature effects were not significant within the range of study (9 - 26'C).
Removal rate, corrected for body size, was similar among all grazer taxa except
amphibians, in which it tended to be lower (Fig. 4.3). Rates calculated from field
experiments were similar among stream and standing water sites but lower than
those from experimental systems. This is most likely to be a result of density-
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
Snails
Caddisflies
-----{---IJ--------
----en-{[J-
Amphibians
Fishes
Mayflies
Chironomids
•
m
~J{ It-
75
-
•
Figure 4.3. Box-and-whisker plots of the residuals from Equation 1 (see text) separated by
taxa. The median value is marked by the central line; 25 and 75 percentile values form the
ends of the box; whiskers delimit the range of the observations except for extreme values,
represented by points. (From Cattaneo and Mousseau, 1995.)
dependent factors limiting the ungrazed biomass in the field, but other factors such
as less active grazing in the presence of predators or alternative food sources may
have played a role.
Community Composition
Herbivore diet depends on attributes of both herbivore (size, motility, morphological specializations for feeding, digestive capabilities) and periphyton (size, ease of
harvest, density, palatability, nutritional value, and perhaps also reproductive traits
and chemical defenses) (Gregory, 1983*). These factors result in a general selectivity by different animals for particular types of periphyton.
Early studies indicated preference for particular types of periphyton by grazers
(Fryer, 1968; Whiteside et aI., 1978; Moore, 1979; Titmus and Badcock, 1981).
The effects depend on the initial periphyton community (Hart, 1981 *; Kohler,
1984*), the nature and size of the grazers (scrapers, for example, are more
devastating than browsers [Mason and Bryant, 1975; Bowker et aI., 1983; Karouna and Fuller, 1992*]), their abundance (Dodds, 1991 *), and the environmental
conditions (Dodds, 1991 *; Poff and Ward, 1992*).
Selection may be made on the basis of growth form (e.g., adnate diatoms vs.
arborescent species [Peer, 1986; Fryer, 1957, 1968, 1974; Fulton, 1988]), on
particular species (Mills and Wyatt, 1974; Botts and Cowell, 1992), or on size of
particle, irrespective of its taxonomy (Patrick, 1970*; Kessler 1981; Pinder
1992 *). The consequences include the development of different, sometimes less
easily grazeable forms, such as coarse filaments (Eichenberger and Schlatter,
1978*; Cattaneo, 1983; Hart, 1985*; Cattaneo and Kalff, 1986; Garland and
Buikema, 1986; Botts, 1993) which may subsequently be grazed by larger animals
such as crayfish or vertebrates (Dickman, 1968; Osborne and McLachlan,
1985), or restriction of the periphyton community to closely adpressed diatoms
(Hann, 1991). Development of less edible cyanophyte growths may occur but is
not a prominent effect.
76
J.I. Jones, B. Moss, and J.O. Young
Communities that persist under intense grazing pressure have taxa that have
high growth or recruitment rates, are buffered from overexploitation by their
growth form (small, flat, or adnate), pass through the guts of the grazer unharmed,
or are rejected because of unpalatability or handling dificulties (Hann, 1991).
Removal of particular algal taxa may alter the assemblage by directly reducing the
dominant algae or by increasing other algal species with poor competitive ability.
The consequences of grazing may differ in subtropical and tropical lakes. Botts
and Cowell (1993) found that abundances of epiphytic algae and invertebrate
grazers were only weakly correlated in a subtropical lake, whereas in temperate
lakes, they usually show strong temporal correlation.
Grazing by medium-sized fauna on filaments is mechanically possible with
long chain diatoms anchored to a substratum but impossible for long filamentous
blue-green and green algae where there is cohesion between strands (Nicotri,
1977). However, large grazers are the most effective at eating filaments. Anuran
larvae eat periphyton (Dickman, 1968; Osborne and McLachlan, 1985) and can
remove massive floating clumps of Mougeotia. Many American fish are algivores
(Scott and Crossman, 1973) or use filamentous algae as a substantial food source,
particularly in late summer (France et aI., 1991). Power and Matthews (1983*) and
Power et al. (1985*) showed strong complementarity between the distribution of a
herbivorous minnow, Campostoma anomaium, and filamentous green algae in
streams. Cattaneo (1983) and Cattaneo and Kalff (1986) excluded minnows,
crayfish, and tadpoles from cages and observed increases in Mougeotia and
Oedogonium over diatoms. Crayfish are opportunistic feeders on animals, plants,
and algal filaments (Flint and Goldman, 1975), and it has been shown they can
dramatically reduce macroalgae (Schindler et aI., 1985; Aloi, 1988; Creed, 1988).
Grazer control on metaphyton flocs or clouds appears to be effective only at the
start of growth. Once the cloud develops, it becomes effectively "immune" and
functions like a macrophyte, with grazers removing the algae attached on its
surfaces (Dodds, 1991 *). Dudley (1986*) found mayflies able to control Cladophora when it was sparse, but they had no effect once it was established. The
often-abundant presence of invertebrates in clouds may be because they act as a
predator refuge rather than a food source, although entrapment of detrital particles
may attract some invertebrates (Dudley et aI., 1986*; Dodds, 1991 *). Even vertebrates may have little effect once the algal mass is well established (France et aI.,
1991) and grazing is opportunistic and not preferential (Tallman et al., 1984).
Productivity
Studies on the effects of grazing by animals other than mollusks on the productivity of periphyton are rare and their results somewhat contradictory. Grazing by
snails and by stream animals has been shown to reduce periphyton productivity
(Rosemond et al., 1993). Other studies have indicated a decrease on an areal basis
but an increase on a biomass-specific basis (Lamberti and Resh, 1983*; Gelwick
and Matthews, 1992*), or an increase of areal primary production (Lamberti et aI.,
1989*; Power, 1990a*).1t would appear that enhancement of production is due to
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
77
nutrient renewal by algal cell disruption and excretion by grazers (Underwood
1991), reduced competition for nutrients (Lamberti and Resh, 1983*), increased
light levels to lower strata of periphyton, and removal of senescent cells by
consumption or dislodgment (Lamberti et aI., 1987a*), but these gains must
outweigh the losses to grazing for effects to be seen. This will depend on both the
intensity of grazing and what factors are limiting algal growth. Hence, Feminella
and Hawkins (1995*) suggest that grazers in streams have a less consistent effect
on productivity than on abundance and biomass.
Disturbance
Periphytic algae may be dislodged or suffer mechanical damage by grazers during
feeding or, in the case of caddisflies and chironomids, case building (Cattaneo,
1983; Pringle, 1985*; Botts, 1993). In streams, this organic matter may be important for browsers (metaphyton), filter feeders, or once settled, deposit-collecting
detrivores (Lamberti et al., 1987a*, 1989*, 1995*). The proportion ofperiphyton
removed in this way may be large (Cattaneo and Mousseau, 1(95).
Differences in the efficiency by which algal taxa are digested can also alter
algal dominance patterns. Certain algae, particularly small diatoms, pass intact
through the grazer gut (Brown, 1960; Peterson, 1987), to be deposited in nutrientrich feces (Vaughn, 1986*).
Nutrient Enhancement due to the Presence of Grazers
Mollusks and nonmolluscan stream grazers can positively affect the supply of
nutrients to the grazer-resistant algal understory (e.g., McCormick and Stevenson,
1991 *; McCormick, 1994*), either by excreting nitrogen directly or by removing
the periphyton overstory and facilitating diffusion from the water column to the
understory. The only nonmolluscan lake study (Flint and Goldman, 1975), however, suggested that increased productivity of periphyton due to the presence of
crayfish was not due to crayfish excretion. On a local scale, motile diatoms may
migrate to chironomid tubes and use excreted nutrients (Botts, 1993).
Periphyton Effects on the Grazers
The quantity, quality, and composition of periphyton can have an important effect
on grazer distribution, abundance, growth rate, life history, and probably, production and community composition. Information on the last two is lacking.
Numerical response of invertebrates to changes in periphyton abundance may be
attributable to migration, emergence, or reproduction.
Distribution
Mason and Bryant (1975) and Kornij6w (1992) found some species of chironomid
larvae moved from mud in late spring as temperatures rose onto plants and grazed
78
J.I. Jones, B. Moss, and 10. Young
periphyton; they returned to the mud in autumn as temperatures fell. Gressens and
Lowe (1994) observed dispersal of chironomid larvae among algal patches differing in abundances of several diatom species and the green alga Stigeoclonium.
Patch preference of the larvae was negatively correlated with abundance of Stigeoclonium, chlorophyll a concentration, and algal biovolume and positively correlated with algal diversity. Periphyton quality was more important than quantity. In
streams, caddisfly and mayfly species respond to variations in periphyton abundance by selectively colonizing and spending more time on patches with more
periphyton (Hart, 1981 *; Feminella and Resh, 1991 *). Selection for algal abundance
was of overriding importance, even though algal type was shown to have marked
effects on the development of the grazers (Vaughn, 1986*). Presumably, selection of
food source will depend on food availability and competition from other grazers.
Abundance
In their review of stream work, Feminella and Hawkins (1995*) found higher
grazer densities at high periphyton abundance (caddisflies, mayflies, and chironomids were the most frequently observed taxa). Those studies that have been
conducted in lakes seem to show a similar situation. Numbers of chydorid cladocemns and chironomids were found to be highly correlated with diatom numbers
(Fairchild, 1981), and in another study, chironomid density increased with an
-,
C\I
E
10
'0
.. /
Q
E
U)
U)
as
E
•
..~.
1
0
L
.a
...
CD
...as
N
Q
••••
.~
••
•
0.1
•
100
1,000
chlorophyll a (JJg dm -2 )
Figure 4.4. Correlation between periphyton biomass collected from sites in Lake Memphremagog, Quebec, and the biomass of oligochaetes and chironomids from the same sites
4-5 days later. (From Cattaneo, 1983.)
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
79
increase in algal biomass (Fairchild and Lowe, 1984). However, there may be a lag
in the response of invertebrates (Fig. 4.4); chironomid larvae, oligochaetes, and
nematodes were most numerous in epiphytic communities living on Equisetum
soon after the peak of algal abundance (Kairesalo, 1984). This may explain reports
of higher grazer density in low-periphyton situations. Herbivore populations appear to be controlled from the bottom up by algal availability, which may occur in
conjunction with strong top-down control of periphyton biomass by grazing.
Growth
No studies on the effects of periphyton on the growth of grazers in lakes have been
made. The review by Feminella and Hawkins (1995*) showed that most studies in
streams found highest growth at high periphyton biomass. Only two studies
reported highest growth in low periphyton biomass.
The nature of the periphyton food source may also influence growth (Lamberti
et ai., 1989*). Assimilation efficiency of 20 aquatic insects, using 45 published
values was 30--60% for periphyton (70-95% for animal food; 5-30% for detritus).
For a snail, Juga silicula, it was 70-80% when the snail was fIrst added to a stream
but declined with time (to 40%). This coincided with a shift in composition from
diatoms and unicellular greens to fIlamentous green and blue··green algae. Variation in protein and lipid content and cell wall thickness among periphyton is likely
to influence palatability. High C:N (> 17: 1) is thought to be deleterious. Periphyton
has CN ratios of 4:1-8:1; cf. 13:1-69:1 in macrophytes (Gregory, 1983*).
In the slow-flowing rivers of southeast England young-of-the-year fIsh (roach,
Rutilus rutilus, and chubb, Leuciscus cephalus) switch to feeding on periphyton
when preferred zooplankton are unavailable, either due to low numbers (Garner,
1996a*) or at night when sight feeding is precluded (Garner, 1996b*). This switch
coincides with a decline in the growth rate of these fish (Garner et ai., 1995*).
Although they are not prominent in many periphyton communities, cyanobacteria
are generally poor food. They contain much protein, but their mucosaccharide sheath
is indigestible and toxins may be produced. This was reflected in a slower development of caddisflies when fed cyanobacteria than other algae (Vaughn, 1986*).
Evidence from grazing experiments is mixed. Gammarns would not consume Phormidium (Moore, 1975), but orthoclad midges have suppressed Phormidium and
Oscillatoria in outdoor channels (Eichenberger and Schlatter, 1978*). Most studies
have been on fIlaments. The situation may be different for colonies or unicells.
Life History
As well as the effects on growth, invertebrate development is influenced by the
food source. The rate at which invertebrates pass through their life cycle is
influenced by both the quality (Vaughn, 1986*) and quantity (our own results,
unpublished) of the available periphyton. There is also a considerable effect on
fecundity (our own results, unpublished), and subsequently population size. It is
not surprising, therefore, that invertebrate life cycles are often timed to coincide
with food availability (Tokeshi, 1986a*).
80
J.I. Jones, B. Moss, and J.O. Young
Effects of Grazers on Other Grazers
Different grazer species may interfere with each other. Intense competition for food
can occur (Tokeshi, 1986b*), which will lead to less selection as food availability is
reduced. The processes of hatching, migration from mud to plants and vice versa,
pupation, cocoon formation, and emergence will affect the species pool of invertebrates. Often these processes are closely coupled with food availability, leading to
populations synchronized with the spring peak ofperiphyton (Tokeshi, 1986a*).
Both inter- and intraspecific competition have been shown to occur between
grazers in lakes. Kajak (1963) found that increased density of Chironomus reduced the densities of snails and other midges, whereas Cattaneo (1983) showed
the reverse, with increases in midges, ostracods, and chydorids in the absence of
snails. Negative effects of snails on chironomids, snails on chydorids, snails on
snails, and chironomids on chironomids have been detailed, inter alia, by Gressens
(1995) (Fig. 4.5) and Cuker (1983). Both positive and negative effects have been
recorded between snails and tadpoles (Bronmark et al., 1991). Snails feed on less
nutritious algae in the presence of tadpoles but encourage algal communites
beneficial to tadpole growth, whereas tadpoles compete with one another for food
o snails
13
~ 20
-- -- --
CIJ
E
CIJ
VI
CIJ
.g
.15
t. . ~·~.~.~il -----------1
10
E
a.
o....
4snails
.05
Q.
E
.c
~
;~
12
························ .............. 0
. .·1-------- -----------t
1 snail
:0
8
; : . . . . +~'~~;I:. . . . . . . . . . . . . . . . j
(J)
:>
o
8
u
....
CIJ
~ct~
d
z
6
4
2
'"
~
_______
2~~~~
____ _
"
4 snails
........................................................... .
o
50
250
Initial no. midges
Figure 4.5. Response of grazers to
variations in midge and snail densities in enclosures in Gas Station
Pond, Illinois. The response of midges
was measured as the proportion of
midge larvae that emerged. Change
in weight of snails indicates the snail
response. The number of Pleuroxus
recovered was used to estimate chydorid populations. (From Gressens,
1995.)
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
81
(Bronmark et al., 1991). Interference with feeding activity and dwelling construction appears to be the predominant form of competition between grazers.
Field exdosure experiments, using different mesh sizes, have shown that the
largest grazers normally dominate (Cattaneo and Kalff, 1986). Small grazers such
as oligochaetes, chironomids, and Cladocera replaced snails in finer-mesh enclosures, producing increased numbers of individuals but a similar total biomass. This
reduction in grazer size but not biomass did not result in reduced algal biomass
(with one exception) but in a dominance by small algae, characterized by a high
turnover rate. A similar situation has been found in laboratory microcosms (Brock
et al., 1995), where chemical removal of arthropod grazers produced only a
short-term increase in periphyton, before snail and oligochaete populations responded to competitive release. Subsequent periphyton biomass was no different
from untreated controls.
Effects of Predators on Grazers
Studies on the importance of cascading trophic interactions (top-down) effects in
freshwater benthic food chains to the periphyton trophic level have focused on fish
and not on invertebrate predators such as triclads, leeches, coleopterans, hemipterans, and odonatans. Moss (1976) found that the addition of bluegill sunfish to
fertilized ponds caused an increase in biomass of certain macrophytes and epiphytes, which was probably due to fish predation on the grazing invertebrates
(amphipods and mayflies). Mazumder et al. (1989) showed that yellow perch was
associated with lower abundances of amphipods and chironomids and higher
concentrations of periphyton particulate phosphorus. In the absence of fish, the
invertebrates were more abundant, and periphyton productivity values were lower
(Fig. 4.6). In experiments in a Swedish eutrophic pond, Bronmark (1994)
concluded that nonmolluscan benthic macroinvertebrates were not greatly
affected by the presence of tench or perch, whereas tench produced a positive
effect on the biomass of periphyton by eating snails and reducing grazing
pressure.
Batzer and Resh (1991) studied interactions among a predaceous hydrophilid
beetle larva, a grazing chironomid larva, and periphyton in 11 experimental ponds.
Abundant beetle larvae reduced autumn densities of chironomid larvae, which
allowed periphyton to grow independently of midge density. In winter, densities of
beetle larvae declined and midge populations increased. At a lower density of
beetles, a close relationship was found between the densities of the chironomid
and periphyton biomass, which was not seen when beetles were abundant
(Fig. 4.7). Grazing by chironomids reduced periphyton biomass, which sometimes
resulted in insufficient food reserves to support high chironomid production.
Although many chironomids were eaten, their seasonal production was slightly
greater when beetles were abundant than when they were few.
Interactions between piscivorous and herbivorous fish can also have consequences for periphyton. The addition ofpiscivores to streams containing the stone
82
J.I. Jones, B. Moss, and J.O. Young
150
--Ie
N
338! 52·8
100
8c
o
"0
.B
c:s:
50
a
III
'0e
c
e
.c.
"0
.c.
Q.
E
c:s:
u
III
1
Q)
Q.
0
u
III
c
0
~
Q)
U
0
"0
0
U
'"
"0
'0
~
"0
>-
r
\/I
c
0
~
0
c:
0
Q
II>
0
l:)
Figure 4.6. Abundance of invertebrates from enclosures in Lake St. George, Ontario,
subjected to four treatments: control (horizontal lines), fish added (solid), nutrients added
(wide hatched), and nutrients and fish added (narrow hatched). (From Mazumder et aI.,
1989.)
roller (Campostoma) and locariid catfish, which are effective removers of
algae, led to ultimate increases in filamentous periphyton (Power and Matthews, 1983*).
In streams, the effects of fish can be pronounced. In the absence of gravel
refuge, selective feeding on invertebrate predators such as damselfly larvae, can
lead to increased populations of small algi vores and hence fewer algae (Power,
1990b*; Power, 1992*).
The most important aspect of the periphyton-grazer-fish interaction in lakes is
that these herbivorous invertebrates provide an alternative food source for fish and
hence reduce the pressure on zooplankton. Without such invertebrates, the ontogenetic shifts seen in the feeding of adult fish would not occur (see Persson and
Crowder, this volume, Chapter 1; Diehl and Kornij6w, this volume, Chapter 2),
and all fish would compete for the same food resource, a situation seen in plantless
eutrophic lakes. Such feeding on invertebrates by fish will result to some extent in
an increase in periphyton, but the overall cost to the plants may be outweighed by
gains resulting from increased zooplankton abundance.
83
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
A. 100. PLANT -COVER
1000
------0
Q
Z
----
0
Il.
........
~
100
8o~
w
2"
0
0
0
0
~
0
0
,
10
1000
100
10
PERIPHYTON IN DECE ..SER ( ..G DRY wr./SAWPLE UNIT)
8. 50. PLANT-COVER
0
0
1000
0
0
----0-
0
z
--D_
0
0
O
---
--
Q..
"-
~
0
0
100
w
0
"a
:i
10
I
10
I
100
I
1000
PERIPHYTON IN DECE"BER ( ..G DRY WT./SAWPLE UNIT)
Figure 4.7. Comparison of periphyton biomass and densities of midge larvae (Cricotopus
sylvestris) from ponds in a manipulated Californian wetland collected over the subsequent
3 months. Hydrophilid larvae were scarce in (A) the 100% plant-covered areas but abundant
in (B) the 50% plant-covered areas. (From Batzer and Resh, 1991.)
84
J.1. Jones, B. Moss, and lO. Young
Overview
There seems no doubt that invertebrate grazers can severely crop periphyton
growths and that, in temperate habitats at least, they are frequently allowed to do
so by predation rates that allow considerable buildup of their populations. This has
been acknowledged for snail populations for some time; the current review suggests that it is no less true for nonmolluscan grazers, particularly the larvae and
nymphs of insects. It is thus possible that epiphytic grazers are as important as
planktonic ones in maintaining conditions appropriate to macrophyte dominance
of the primary production in shallow lakes.
Much of the literature centers on individual species or restricted groups, however, and there is a shortage of studies of whole communities, containing more
than two trophic levels. Occasional references to instances when the periphyton is
not controlled by invertebrate grazers suggests that, sometimes, predation on these
grazers may be considerable, but an overall picture of the importance of such
relationships, as is available for the temperate plankton, is not yet possible. The
fact that grazing rates measured in experimental enclosures from which predators
are deliberately or inadvertently excluded are greater than in the open habitat also
suggests an important role for top-down effects.
In continental areas with rich vertebrate faunas, there may be an important role
for vertebrate grazers on periphyton, including both fish and amphibians. There do
not appear to be specific periphyton grazers among birds or mammals. Although
such vertebrates frequently eat macrophytes and hence the associated periphyton,
their essentially aerial adaptations do not allow the delicate positioning that is
needed to graze the periphyton without damaging the host plant. Experimental
work on vertebrate periphyton grazers is scarce; they appear to be particularly
important in grazing the coarse filamentous algae that may form large blanketing
flocs or clouds in the water. Small invertebrate grazers appear to be able to tackle
these only when the flocs are sparse but can prevent their development if grazing
sets in early enough. Such filamentous flocs may be more damaging to macrophyte growth than thin adnate layers of periphyton, and the gap in the literature
concerning them and their vertebrate grazers is an important one.
Currently, the literature is dominated by studies on periphyton films on rocks
and stones in streams and, in standing waters, by work on mollusks. Grazing rates
in streams and standing waters appear broadly similar, as do those of snails and
other invertebrates. It is not yet clear whether it matters much that the periphyton substrate is living or inert. Comprehensive work (our unpublished data)
suggests that living substrates only minimally influence the potential size of
the periphyton community; the extent to which its composition may be determined by the activities of the host is still not fully resolved, but the complexities may be far less great than hitherto thought. The architecture and
growth rate of the macrophyte, rather than any chemical influence, may be
most important, but determination of the problem is confounded by random
effects of colonization into the periphyton community, coupled with selective
effects of grazing after colonization.
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
85
Such selective effects undoubtedly occur, but they are not as simple as they
appear to be in the plankton, where small species are more vulnerable than larger
ones. Some periphyton grazers take large cells, others small cells. The effect on a
previously ungrazed periphyton community will vary, dependent on the nature of
the grazers present. In tum, this will depend on a large range of factors, including
the architecture of the host plant, the degree of competition, the time of year and
the phasing of different life histories, the action of predators, the vagaries of
colonization, and probably other factors. Again, this is a very different situation to
that in the plankton, in which the activities of a limited range of grazing animals,
of similar size range and grazing methods (filter and suspension feeding), effectively hunted by one major sort of predator, are concerned. The possibilities of
predicting and modeling the outcome of events in the plankton are much greater
than are yet possible in periphyton grazer systems.
This review has highlighted the urgent need for research, both in the field and
laboratory, due to the paucity of information concerning lacustrine nonmolluscan
groups in the following areas: ingestion and digestion rates; assimilation efficiencies; effects of grazer type (including mouthpart morphology), selectivity, size,
density, multiple species action, disturbance, and nutrient enhancement on individual species and whole communities of periphyton including standing crop,
production, and composition; the effect of periphyton on grazer abundance, distribution, growth rates, production, and community composition; interrelationships between grazing species; and the effect of predation on grazers, and hence
periphyton.
Overall, the macrophyte-periphyton-grazer-predator system offers considerable parallels to that well studied on a larger scale by range managers in prairies
and savannahs. Rather similar elements are present. Frequently, we talk of the
periphyton in terms merely of its chlorophyll, its grazers as numbers or biomass.
No one would talk of an African savannah as a uniform growth of grass with a
single sort of grazer, say, an antelope. The plant communities vary from long
grasses, after the rains, on which several species of large antelope may feed to
short grasses supporting small antelope, where there has been previous heavy
grazing. A variety of rodents feeds on seeds, whereas grasshoppers and locusts
also graze the fresh grasses, and a huge biomass of termites takes the dead culms
and fragments of wood from thorn bushes and shrubs. Some of the grazers,
browsing as individuals, also depend on these (e.g., the black rhinoceros). Others
gather in herds (e.g., elephants), temporarily to devastate the vegetation before
moving on and allowing a new succession to begin with a new set of grazers. The
plant community will vary in species composition, dependent on slope and soil,
previous history, and microclimate; a parallel array of specialist insects will graze
the different communities. Streams crossing the landscape will bear riverine
gallery forest with a completely different array of grazers, including smaller
antelopes such as the puku and perhaps hippopotami in prominent pools.
If this seems like an unreasonable comparison with the periphyton system, then
consider the scale. A periphyton diatom might be 5 x 5 x 30 /lm in size. If it be
considered equivalent to a clump of grass of base 5 cm x 5 cm, a square centimeter
86
1.1. Jones, B. Moss, and J.O. Young
of aquatic plant surface is equivalent to a hectare of land surface and a whole
aquatic plant to several kilometers squared. It is by adjusting our thinking to the
appropriate scale that we may best understand the functioning and importance of
the periphyton system.
References
Aloi, J.E. Changes in periphyton biomass, primary productivity and community composition resulting from field manipulations of crayfish. Bull. North Am. Benth. Soc. 5:103;
1988.
Aloi, J.E. A critical review of recent freshwater periphyton field methods. Can. J. Fish.
Aquat. Sci. 47:656-670; 1990.
Batzer, D.P.; Resh, V.H. Trophic interactions among a beetle predator, a chironomid grazer,
and periphyton in a seasonal wetland. Oikos 60:251-257; 1991.
Bergey, E.A Local effects of a sedentary grazer on stream algae. Freshwat. BioI. 33:401409; 1995.
Botts, P.S. The impact of small chironomid grazers on epiphytic algal abundance and
dispersion. Freshwat. BioI. 30:25-33; 1993.
Botts, P.S.; Cowell, B.C. Feeding electivity of two epiphytic chironomids in a subtropical
lake. Oecologia 89:331-337; 1992.
Botts, P.S.; Cowell, B.C. Temporal patterns of abundance of epiphytic invertebrates on
Typha shoots in a subtropical lake. J. North Am. Benth. Soc. 12:27-39; 1993.
Bowker, D.W.; Warehom, M.T.; Learner, M.A The selection and ingestion of epilithic
algae by Nais elinguis (Oligochaeta: Naididae). Hydrobiologia 98: 171-178; 1983.
Brock, T.C.M.; Roijackers, R.M.M.; Rollon, R.; Bransen, E; Van der Heyden, L. Effects of
nutrient loading and insecticide application on the ecology of Elodea-dominated freshwater microcosms. II. Responses of macrophytes, periphyton and macroinvertebrate
grazers. Arch. Hydrobiol. 134:53-74; 1995.
Bronmark, C. Effects of tench and perch on interactions in a freshwater, benthic food chain.
Ecology 75:1818-1828; 1994.
Bronmark, c.; Rundle, S.D.; Earlandsson, A. Interactions between freshwater snails and
tadpoles: competition and facilitation. Oecologia 87:8-18; 1991.
Brown, D.S. The ingestion and digestion of Cloeon dipterum L. Hydrobiologia 16:81-96;
1960.
Cattaneo, A. Grazing on epiphytes. Limnol. Oceanogr. 28:124-132; 1983.
Cattaneo, A., The effect offetch on periphyton variation. Hydrobiologia 206: 1-10; 1990.
Cattaneo, A; Kalff, J. The effect of grazer site manipulation on periphyton communities.
Oecologia 69:612-617; 1986.
Cattaneo, A; Mousseau, B. Empirical analysis of the removal rate ofperiphyton by grazers.
Oecologia 103:249-254; 1995.
Creed, R.P. The influence of crayfish grazing on benthic community structure in a Michigan
stream. Bull. North Am. Benth. Soc. 5:67; 1988.
Cuker, B.E. Competition and coexistence among the grazing snail Lymnaea, Chironomidae,
and Microcrustacea in an arctic epilithic lacustrine community. Ecology 64: 10-15;
1983.
Dickman, M. The effect of grazing by tadpoles on the structure of a periphyton community.
Ecology 49:1188-1190; 1968.
Dodds, W.K. Community interactions between the filamentous alga Cladophora glomerata
(L.) Kuetzing, its epiphytes, and epiphyte grazers. Oecologia 85:572-580; 1991.
Dudley, T. Beneficial effects of grazers on algal growth. Bull. North Am. Benth. Soc. 3:87;
1986.
Dudley, T.L.; Cooper, S.D.; Hemphill, N. Effects of macroalgae in a stream invertebrate
community. J. North Am. Benth. Soc. 3:93-106; 1986.
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
87
Eichenberger, E.; Schlatter, E The effect of herbivorous insects on the production of benthic
algal vegetation in outdoor channels. Verh. Int. Verein. Limnol. 20: 1806-1810; 1978.
Fairchild, G.W. Movement and distribution of Sida crystallina and other littoral microcrustacea. Ecology 62:1341-1352; 1981.
Fairchild, G.W.; Lowe, RL. Artificial substrates that release nutrients: effects on periphyton
and invertebrate succession. Hydrobiologia 114:29-37; 1984.
Fairchild, G.W.; Campbell, J.M.; Lowe, R.L. Numerical response ofChydorids (Cladocera)
and Chironomids (Diptera) to nutrient-enhanced periphyton growth. Arch. Hydrobiol.
114:369-382; 1989a.
Fairchild, G.w.; Sherman, J.w.; Acker, EW. Effects of nutrient (N,P,C) enrichment, grazing
and depth upon littoral periphyton of a softwater lake. Hydrobiologia 173:69--83;
1989b.
Feminella, J.W.; Hawkins, C.P. Interactions between stream herbivores and periphyton: a
quantitative analysis of past experiments. J. North Am. Benth. Soc. 14:465-509; 1995.
Feminella, J.W.; Resh, V.H. Herbivorous caddisflies, macro algae, and epilithic microalgae:
dynamic interactions in a stream grazing system. Oecologia 87:247-256; 1991.
Flint, R.W.; Goldman, C.R. The effects of a benthic grazer on the primary productivity of
the littoral zone of Lake Tahoe. Limnol. Oceano gr. 20:935-944; 1975.
Forsberg, C.; Kleiven, S.; Willen, T. Absence of allelopathic effects of Chara on phytoplankton in situ. Aquat. Bot. 38:289-294; 1990.
France, RL.; Howell, E.T.; Paterson, MJ.; Welbourn, P.M. Relationship between littoral
grazers and metaphytic algae in five softwater lakes. Hydrobiologia 220:9-27; 1991.
Fryer, G. The feeding mechanism of some freshwater cyclopoid copepods. Proc. Zool. Soc.
Lond. 129:1-25; 1957.
Fryer, G. Evolution and adaptive radiation in the Chydoridae (Crustacea: Cladocera): a
study in comparative functional morphology and ecology. Philos. Trans. R. Soc. Lond.
B 254:221-385; 1968.
Fryer, G. Evolution and adaptive radiation in the Macrothricidae (Crustacea: Cladocera): a
study in comparative functional morphology and ecology. Philos. Trans. R Soc. Lond.
B 269:237-274; 1974.
Fulton, RS. III. Grazing on filamentous algae by herbivorous zooplankton. Freshwat. BioI.
20:263-271; 1988.
Garland, J.L.; Buikema, A.L. Effects of meiofaunal grazing on detrital and epiphytic
assemblages. Bull. North Am. Benth. Soc. 3:86; 1986.
Garner, P.; Bass, J.A.B.; Collett, G.D. The effect of weed cutting upon the biota of a large
regulated river. Aquatic conservation: marine and freshwater ecosystems. 6:21-29;
1995.
Garner, P. Microhabitat use and diet of 0+ cyprinid fishes in a lentic regulated reach of the
River Great Ouse. J. Fish BioI. 48:367-382; 1996a.
Garner, P. Diel patterns in the feeding and habitat use of 0 -group fishes in a regulated river
the River Great Ouse, England. Ecol. Freshwat. Fish. 5:175-182; 1996b.
Gelwick, EP.; Matthews, W.l. Effects of an algivorous minnow on temperate stream
ecosystem properties. Ecology 72:1630-1645, 1992.
Gregory, S.v. Plant-herbivore interactions in stream systems. In: Barnes, J.R.; Minshall,
G.W., eds. Stream ecology. New York: Plenum Press; 1983:155-187.
Gressens, S.E. Grazer density, competition and the response of the periphyton community.
Oikos 73:336-346; 1995.
Gressens, S.E.; Lowe, R.L. Periphyton patch preference in grazing chironomid larvae. J.
North Am. Benth. Soc. 13:89-99; 1994.
Hann, B.l. Invertebrate grazer-periphyton interactions in a eutrophic marsh pond. Freshwat. BioI. 26:87-96; 1991.
Hart, D.D. Foraging and resource patchiness: field experiments with a grazing stream
insect. Oikos 37:46-52; 1981.
88
J.1. Jones, B. Moss, and J.O. Young
Hart, D.D. Grazing insects mediate algal interactions in a stream benthic community. Oikos
44:40-46; 1985.
Irvine, K.; Moss, B.; Bales, M.; Snook, D. The changing ecosystem of a shallow, brackish
lake, Hickling Broad, Norfolk, U.K. I. Trophic relationships with special reference to the
role of Neomysis integer. Freshwat. BioI. 29:119-139; 1993.
Jacoby, J.M. Grazing effects on periphyton by Theodoxus fluviatilis (Gastropoda) in a
lowland stream. J. Freshwat. Ecol. 3:265-274; 1985.
Jacoby, J.M. Alterations in periphyton characteristics due to grazing in a Cascade foothill
stream. Freshwat. BioI. 18:495-508; 1987.
Kairesalo, T. The seasonal succession of epiphytic communities within an Equisetum
fluviatile L. stand in Lake Paajarvi, Southern Finland. Internationale Rev. Ges. HydrobioI. 69:475-505; 1984.
Kairesalo, T.; Koskimies, I. Grazing by oligochaetes and snails on epiphytes. Freshwat.
BioI. 17:317-324; 1987.
Kajak, Z. The effect of experimentaly induced variations in the abundance of Tendipes
plumosus L. larvae on intraspecific and interspecific relations. Ekol. Pol. 11 :355-367;
1963.
Karouna, N.K.; Fuller, RL. Influence of four grazers on periphyton communities associated
with clay tiles and leaves. Hydrobiologia 245:53-64; 1992.
Kessler, D.H. Grazing rate determination of Corynoneura scutellata Winnertz (Chironomidae: Diptera). Hydrobiologia 80:63-66; 1981.
Kohler, S.L. Search mechanism for a stream grazer in patchy environments: the role offood
abundance. Oecologia (Berlin) 62:209-218; 1984.
Knudson, B.M. Ecology of the epiphytic diatom Tabellaria flocculosa (Roth) Kutz. var.
flocculosa in three English lakes. J. Ecol. 45:93-112; 1957.
Kornij6w, R Seasonal migration by larvae of an epiphytic chironomid. Freshwat. BioI.
27:85-89; 1992.
Lamberti, G.A; Resh, V.H. Stream periphyton and insect herbivores: an experimental study
of grazing by a caddisfly population. Ecology 64:75-81; 1983.
Lamberti, G.A; Moore, J.W. Aquatic insects as primary consumers. In: Resh, Y.H.; Rosenberg, D.M., eds. The ecology of aquatic insects. New York: Praeger; 1984: 164-195.
Lamberti, G.A.; Feminella, J.w.; Resh, Y.H. Herbivory and intraspecific competition in a
stream caddisfly population. Oecologia 73:75-81; 1987a.
Lamberti, G.A; Ashkenas, L.R; Gregory, S.Y.; Steinman, AD. Effects of three herbivores
on periphyton communities in laboratory streams. J. North Am. Benth. Soc. 6:92-104;
1987b.
Lamberti, G.A; Gregory, S.Y.; Ashkenas, L.R.; Steinman, AD.; McIntire, C.D. Productive
capacity of periphyton as a determinant of plant-herbivore interactions in streams.
Ecology 70:1840-1856; 1989.
Lamberti, G.G.; Gregory, S.V.; Ashkenas, L.R.; Li, J.L.; Steinman, AD.; McIntire, C.D.
Influence of grazer type and abundance on plant-herbivore interactions in streams.
Hydrobiologia 306:237-247; 1995.
Mason, c.F.; Bryant, RJ. Periphyton production and grazing by chironomids in Alderfen
broad, Norfolk. Freshwat. BioI. 5:271-277; 1975.
Mazumder, A.; Taylor, W.D.; McQueen, D.I.; Lean, D.RS. Effects of nutrients and grazers
on periphyton phosphorus in lake enclosures. Freshwat. BioI. 22:405-415; 1989.
McCormick, P.V. Lotic protistan herbivore selectivity and its potential impact on benthic
algal assemblages. J. North Am. Benth. Soc. 10:238-250; 1991.
McCormick, P.V. Evaluating the multiple mechanisms underlying herbivore-algal interactions in streams. Hydrobiologia 291 :47-59; 1994.
McCormick, P.V.; Stevenson, R.I. Grazer control of nutrient availability in the periphyton.
Oecologia 86:287-291; 1991.
Mills, D.H.; Wyatt, J.T. Ostracod reactions to non-toxic and toxic algae. Oecologia 17: 171177; 1974.
4. Periphyton-Nonmolluscan Invertebrate-Fish Interactions
89
Mitchell, S.E; Wass, RT. Quantifying herbivory: grazing consumption and interaction
strength. Oikos 76:573-576; 1996.
Moore, J.W. The role of algae in the diet of Asellus aquaticus L. and Gammarus pulex L. J.
Anim. Ecol. 44:719-730; 1975.
Moore, J.W. Factors influencing algal consumption and feeding rate in Heterotrissocladius
changi Saether and Polypedilum nubeculosum. Oecologia 40:219-227; 1979.
Moss, B. The effects of fertilization and fish on community structure and biomass of aquatic
macrophytes and epiphytic algal populations: an ecosystem experiment. J. Ecol. 64:
313-342; 1976.
Nicotri, M.E. Grazing effects of four marine intertidal herbivores on the microflora. Ecology 58:1020-1032; 1977.
Osborne, P.L.; McLachlan, A.J. The effect of tadpoles on algal growth in temporary,
rain-filled rock pools. Freshwat. BioI. 15:77-87; 1985.
Patrick, R Benthic stream communities. Am. Scientist 58:546-549; 1970.
Patterson, D.M.; Wright, S.J.Z. The epiphylous algal colonization of Elodea canadensis
Michx.: community structure and development. N. Phytol. 103:809-819; 1986.
Peer, RL. The effects of microcrustaceans on succession and diversity of an algal microcosm community. Oecologia 68:308-314; 1986.
Peterson, G.P' Gut passage and insect grazer selectivity of lotic diatoms. Freshwat. BioI.
18:461-468; 1987.
Pinder, L.C.V. Biology of epiphytic Chironomidae (Diptera: Nematocera) in chalk streams.
Hydrobiologia 248:39-51; 1992.
Poff, N.L.; Ward, J.Y. Heterogeneous currents and algal resources mediate in situ foraging
activity of a mobile stream grazer. Oikos 65:465-478; 1992.
Power, M.E. Resource enhancement by indirect effects of grazers: armored catfish, algae,
and sediment. Ecology 71:897-904; 1990a.
Power, M.E. Effects offish in river food webs. Science 250:811-814; 1990b.
Power, M.E. Habitat heterogeneity and the functional significance of fish in river food
webs. Ecology 73:1675-1688; 1992.
Power, M.E.; Matthews, W.J. Algae-grazing minnows (Campostoma anomulum), pi scivorous bass (Micropterus spp.), and the distribution of attached algae in a prairie-margin
stream. Oecologia 60:328-332; 1983.
Power, M.E.; Matthews, W.J.; Stewart, A.J. Grazing minnows, piscivorous bass, and stream
algae: dynamics of a strong interaction. Ecology 66:1448-1456; 1985.
Prejs, K. Bottom fauna. In: Pieczynska, E., ed. Selected problems of lake littoral ecology.
Warsaw: University of Warsaw; 1976:123-144.
Pringle, C.M. Effects of chironomid (Insecta: Diptera) tube-building activities on stream
diatom communities. J. Phycol. 21:185-194; 1985.
Rosemond, A.D.; Mulholland, P.J.; Elwood, J.W. Top-down and bottom-up control on
stream periphyton: effects of nutrients and herbivores. Ecology 74:1264-1280; 1993.
Sarvala, J.; I1mavirta, Y.; Paasivirta, L.; Salonen, K. The ecosystem of the oligotrophic lake
Paajarvi. 3. Secondary production and the ecological energy budget of the lake. Verh.
Int. Verein. Limnol. 21 :454-459; 1981.
Schindler, D. w.; Mills, K.H.; Malley, D.E; Findlay, D.L.; Shearer, J.A.; Davies, 1.1.; Turner,
M.A.; Linsey, G.A.; Cruickshank, D.R. Long term ecosystem stress: the effects of years
of exprimental acidification on a small lake. Science 228: 1395-1401; 1985.
Scimgeour, G.J.; Culp, J.M.; Bothwell, M.L.; Wrona, EJ.; McKee, M.H. Mechanisms of
algal patch depletion: importance of consumptive and nonconsumptive losses in mayfly
diatom systems. Oecologia 85:343-348; 1991.
Scott, W.B.; Crossman, E.J. Freshwater fishes of Canada. Bull. Fish. Res. Bd. Can. 184:1966; 1973.
Seale, D.B. Influence of amphibian larvae on primary production, nutrient flux, and competition in a pond ecosystem. Ecology 61:1531-1550; 1980.
90
J.r. Jones, B. Moss, and 1.0. Young
Sleigh, M.A.; Baldock, B.M.; Baker, J.H. Protozoan communities in chalk streams. Hydrobiologia 248:53-64; 1992.
Sozska, GJ. Ecological relations between invertebrates and submerged macrophytes in the
lake littoral. Ekol. Pol. 23:393-415; 1975.
Steinman, A.D.; Mcintire, C.D. The effects of current velocity and light energy on the
structure of periphyton assemblages in laboratory streams. J. Phycol. 22:352-361; 1986.
Steinman, A.D.; McIntire, C.D. Effects of irradiance on the community structure and
biomass of algal assemblages in laboratory streams. Can. 1. Fish. Aquat. Sci. 44:16401648; 1987.
Steinman, A.D.; Mcintire, C.D.; Gregory, S.Y.; Lamberti, G.A.; Ashkenas, L.R. Effects of
herbivore type and density on taxonomic structure and physiognomy of algal assemblages in laboratory streams. J. North Am. Benth. Soc. 6:175-188; 1987.
Sumner, W.T.; Mcintire, C.D. Grazer-periphyton interactions in laboratory streams. Arch.
Hydrobiol. 93:135-157; 1982.
Tallman, R.E; Mills, K.H.; Rotter, R.G. The comparative ecology of pearl dace (Semotilus
margarita) and fathead minnow (Pimephales promelas) in Lake 114, the Experimental
Lakes Area, northwestern Ontario, with an appended key to the cyprinids of the Experimental Lakes Area. Can. Manage. Rep. Fish. Aquat. Sci. 1756:1-27; 1984.
Titmus, G.; Badcock, R.M. Distribution and feeding of larval chironomidae in a gravel-pit
lake. Freshwat. BioI. 11 :263-271; 1981.
Tokeshi, M. Population dynamics, life histories and species richness in an epiphytic chironomid
community. Freshwat BioI. 16:431-441; 1986a.
Tokeshi, M. Resource utilization, overlap and temporal community dynamics: a null model
analysis of an epiphytic chironomid community. J. Anim. Ecol. 55:491-506; 1986b.
Underwood, GJ.C. Growth enhancement of the macrophyte Ceratophyllum demersum in
the presence of the snail Planorbis planorbis: the effect of grazing and chemical
conditioning. Freshwat. BioI. 26:325-334; 1991.
Vaughn, C.C. The role of periphyton abundance and quality in the microdistribution of a
stream grazer, Helicopsyche borealis (Trichoptera: Helicopsychidae). Freshwat. BioI.
16:485-493; 1986.
Whiteside, M.e.; Williams, J.B.; White, c.P. Seasonal abundance and pattern of chydorid
Cladocera in mud and vegetative habitats. Ecology 59:1177-1188; 1978.
Young, O.w. A limnological investigation of periphyton in Douglas Lake, Michigan. Trans.
Am. Microsc. Soc. 64:1-20; 1945.
5. Impact of Submerged Macrophytes on
Fish-Zooplankton Interactions in Lakes
Erik Jeppesen, Torben L. Lauridsen, Timo Kairesalo, and
Martin R. Perrow
Introduction
Fish have a major structuring impact on the zooplankton communities in lakes
(Hrbacek et ai., 1961; Brooks and Dodson, 1965) that may cascade to the lower
trophic levels and chemical environment (Carpenter et aI., 1985; Carpenter and
Kitchell, 1993). Ample evidence is available from enclosure experiments (e.g.,
Christoffersen et ai., 1993), whole-lake experiments (e.g., Shapiro et aI., 1975;
Benndorf, 1987; Gulati et aI., 1990; Carpenter and Kitchell, 1993), and empirical
analyses (Jeppesen et ai., 1990, 1997). More recently, it has become evident that
0+ fish may play a key role in zooplankton population dynamics (Cryer et aI.,
1986; Mills et aI., 1987), and some studies suggest that fish larvae are responsible
for the midsummer decline in zooplankton (Luecke et ai., 1990; Jeppesen et aI.,
1997), a phenomenon that is often attributed to increased density of inedible
phytoplankton such as cyanobacteria (e.g., De Bernardi and Guisanni, 1990).
Whole-lake (S~mdergaard et ai., 1997) and enclosure (He and Wright, 1992)
experiments support the structuring role of 0+ fish. How the impDrtance of top-down
control of zooplankton by fish varies along a trophic gradient is debated extensively. McQueen et ai. (1986) suggested that the cascading effect of zooplanktivorous fish is stronger in oligotrophic lakes than in eutrophic lakes, but a
growing body of literature argues that the cascading effect of fish is greater in
eutrophic and hypertrophic lakes with respect to the food web in the classic sense
91
92
E. Jeppesen et al.
(Jeppesen et aI., 1990, 1997; Leibold, 1990; Samelle 1992) and the microbial web
(Jeppesen et aI., 1992; JUrgens 1994; JUrgens and Jeppesen, this volume, Chapter 16).
In addition to nutrient-dependent fish effects, the role of fish seems also to
change with lake depth. A cross-analysis of data and existing empirical relations
suggests that plankti-benthivorous fish have a higher impact on the zooplankton in
macrophyte-free shallow lakes than in corresponding deep lakes (Jeppesen et aI.,
in press). This is because fish biomass per unit of area at any given nutrient level
does not change with mean depth (Hanson and Leggett, 1982; Downing et aI.,
1990), implying that the biomass per unit of volume, and thus probably also the
predation on zooplankton, increases with decreasing mean depth. In addition,
vertical migration as an antipredator defense mechanism (Lampert, 1993) is less
effective in shallow lakes. Furthermore, the availability of alternative food sources
for the fish, such as sediment and bottom fauna, is higher in shallow lakes than in
deep lakes because of detritus and thus of higher quality for the benthos of shallow
lakes. In lakes with low oxygen in the hypolimnion, parts of the sediment may
furthermore be inaccessible to the fish. Consequently, the shallow lake zooplanktivorous fish biomass may be sustained at high levels by additional alternative food sources, and they can therefore maintain a higher predation pressure on
zooplankton than in deep lakes (Jeppesen et aI., 1997). In shallow lakes, however,
submerged macrophytes may cover large areas, and if abundant, this may alter the
interaction strength between fish and pelagic zooplankton as zooplankton may use
macrophytes as refuge against predation from fish. The interactions are, however,
complex, as the presence of macrophytes also influences the mutual interaction
between piscivorous fish and prey fish. For instance, small planktivores use macrophytes as an anti predator defense mechanism. Macrophytes may also influence the
competition between various predatory species and, at the juvenile stages, competition
between prey fish and predators (see Persson and Crowder, this volume, Chapter 1).
Yet, we far from fully understand how the interactions between fish and zooplankton
are influenced by macrophytes. However, certain patterns are emerging.
We first briefly describe the zooplankton communities in macrophyte beds.
Thereafter, we discuss how macrophytes may influence the interactions between
fish and zooplankton and show how these interactions may have cascading effects
on phytoplankton, protozoans, and bacterioplankton. Then, we present a tentative
model describing how fish-zooplankton interactions may alter along a nutrient
gradient, and finally, we suggest future research needs. We intend to highlight key
issues emerging particularly from our own results rather than present a full
literature review. In this chapter, the term littoral zone does not only comprise the
nearshore areas but all areas with plants and nearby open water. Thus, in very
shallow lakes the littoral zone may extend to the lake center.
Zooplankton Community in Macrophyte Beds
Crustacean and rotifer communities in plant beds consist of epiphytic, benthic, and
pelagic forms. Most plant-associated species in the littoral zone can be categorized
5. Fish-Zooplankton Interactions
93
as scrapers on solid substrate (e.g., Eurycercus), suckers (some rotifers), or
graspers (cyclopoid copepods) (Gliwicz and Rybak, 1976). Some forms such as
Sida collect seston by filtration while fixed to the plants, but they do show a
certain plasticity, as they periodically appear free-swimming, especially at night
(Vuille, 1991). Others are probably facultative filter feeders (e.g., Chydorus,
Eurycercus, and Acroperus), and others are predators (Polyphemus). The ecological role of these plant-associated cladocerans is, however, virtually unknown.
Most, if not all, pelagic zooplankton species may periodically occur in the plant
beds. It is characteristic that if a littoral zone is present, cladocerans such as
Ceriodaphnia spp., Chydorus sphaericus, Diaphanosoma brachyurum, and cyclopoid copepods are often more abundant here than in the pelagic zone, whereas
rotifer and calanoid copepod densities show the opposite pattern (Gliwicz and
Rybak, 1976; Yuille, 1991; Lauridsen et aI., 1996). In particular, Ceriodaphnia
spp. and D. brachyurum seem well adapted to plant beds, probably because they
are efficient microfiltrators (of bacterioplankton, etc.) (Gliwicz and Rybak, 1976;
DeMott, 1986). In addition, Ceriodaphnia tolerates low-oxygen conditions (Gliwicz and Rybak, 1976).
The study by Jeppesen et aI. (unpublished results) illustrates how habitat choice
depends on plant density. In fish-free enclosures in which zooplankton could
select between open water and three different plant densities (artificial plastic
plants), major differences in relative zooplankton composition and total abundance were found. Bosnuna longirostris preferred open water. Daphnia spp. were
widespread at all plant densities and in open water. D. brachyurum abundance was
highest at intermediate plant volume infested (PVI), and Ceriodaphnia spp.
and cyclopoid copepods abundance was highest in the most dense vegetation
(Fig. 5.1). In accordance with these results, a cross-analysis of data from 13 lakes
conducted by Cyr and Downing (1988) revealed a negative relationship between
the abundance of B. longirostris and plant density but a positive one with the
density or biomass of cyclopoid copepods and other cladocerans. Other studies,
however, showed high abundance of B. longirostris in the vegetation (Pennak,
1966; Gliwicz and Rybak, 1976; Jakobsen and Johnsen, 1987; Lauridsen et aI.,
1996). These differences may reflect variations in fish predation pressure (see
below). The relative contribution of pelagic zooplankton may also depend on plant
bed size, being lower in large beds (Lauridsen et aI., 1996) as pelagic zooplankton
often aggregate in the transitional zone between plant beds and open water
(Lauridsen and Buenk, 1996). In addition. light intensity and water currents may
influence the zooplankton distribution (e.g., Kairesalo and Penttila, 1990). In
accordance with Figure 5.1, the abundance (Cyr and Downing, 1988; Paterson,
1993; Phillips et al., 1996) and biomass (Cyr and Downing, 1988) of microcrustaceans often increase with plant biomass but to a varying degree depending
on plant type (Paterson, 1993). In the study by Paterson (1993), the number of
microcrustaceans per unit of area was an order of magnitude higher in the plant
beds (0.5 x 106 m 2) than in open water and plant-free sediment in the littoral zone.
Likewise, the average size of the different species is often larger inside the
vegetation than outside (Vuille and Maurer, 1991; Lauridsen and Buenk, 1996).
94
E. Jeppesen et a!.
o
300
600
900
1200
1500
1800
Total
20
Contribution (%)
40
60
eo
100
80smina
Daphnia
Diaphanosoma
Cyclopoid
cope pods
Figure 5.1. (Upper panel) Total abundance and (lower panel) percentage distribution in
terms of numbers of various zooplankton in fish-free enclosures with contrasting densities
of artificial plants (ivy imitations) (0, 10,40, and 80% plant volume infested [PVI]) in Lake
Stigsholm July 1995. The data represent a day-night average. (Modified from Jeppesen et
a!., in preparation.)
Increasing invertebrate density with increasing plant density and high densities per
square meter in the plant beds have also been observed in the case of macro invertebrates (Cyr and Downing, 1988; Diehl and Komij6w, this volume, Chapter 2).
Diel Horizontal and Littoral Vertical Migration of Zooplankton
The presence of planktivorous fish in both the pelagic and the littoral zone may
alter the horizontal distribution of zooplankton. Some recent studies show daytime
aggregation of pelagic zooplankton in the littoral zone (Lauridsen and Buenk,
1996; Stansfield et aI., 1997), whereas other studies support the "shore avoidance hypothesis" proposed by Hutchinson (1967). These differences seem to be
highly dependent on season, reflecting differences in density and horizontal distribution offish. Cryer and Townsend (1988) studied the horizontal distribution of
pelagic zooplankton during 2 years with contrasting densities of 0+ fish and found
that in July densities of Daphnia hyalina and B. longirostris were an order of
magnitude higher in the pelagic than the littoral zone in the low-fish year but up to
lOO-fold more abundant in the littoral than in the pelagic zone when fish density
was high (D. hyalina), Likewise, Jakobsen and Johnson (1987) recorded an even
5. Fish-Zooplankton Interactions
95
distribution in the pelagic and littoral zones of Daphnia longispina and Bosmina
longispina in Lake Kvernavatn, Norway, early in the season, but a segregation
took place when sticklebacks aggregated in the littoral zone, D. longispina moving
to the pelagic zone, while B. longispina, being smaller and less predation vulnerable, moved to the littoral zone.
The horizontal distribution of zooplankton also varies from day to night. Based
on zooplankton studies in shallow Hoveton Great Broad, Timms and Moss (1984)
proposed that pelagic zooplankton move into macrophyte beds during daytime,
using it as spatial refuge against fish predation, but migrate into open water at
night to feed on phytoplankton. Since this study, the existence of diel horizontal
migration (DHM) has been confirmed by several studies (Davies, 1985; De
Meester et aI., 1993; De Stasio, 1993; Lauridsen and Buenk, 1996; Lauridsen et
aI., 1996; Jeppesen et al., 1992; Lauridsen et aI., this volume, Chapter 13). Diel
and seasonal studies undertaken in the littoral zone by Kairesalo (1980), Lehtovaara and Sarvala (1984), Walls et al. (1990), and Stansfield et al. (1992) lend
further support to DHM. Studies undertaken in several lakes with varying fish
densities suggest that the extent of DHM is positively related to the density of
planktivorous fish in the pelagic zone (Lauridsen et aI., this volume, Chapter 13).
Likewise, fish density seems to decide the migrating species and size classes. In
Lake Ring, Denmark, where the abundance of planktivorous fish was low, Lauridsen and Buenk (1996) thus found that Daphnia magna showed higher levels of
DHM than the smaller Daphnia galeata. In the more fish-rich Lake Stigsholm,
even small species such as B. longirostris and Ceriodaphnia spp. migrated (Lauridsen et al., 1996; Jeppesen et al., submitted) (Fig. 5.2). In addition, other studies
have shown that it is the largest individuals of the various species in particular that
undergo DHM (Lauridsen and Buenk, 1996; Pedersen, unpublished results). All
these observations are in accordance with the size-efficiency hypothesis (Brooks
and Dodson, 1965). The empirical evidence offish-mediated horizontal migration
is confirmed by controlled laboratory and field experiments. Laboratory experiments by Lauridsen and Lodge (1996) showed that the presence of fish (Lepomis
eyanellus) or kairiomone cue in open water initiated a migration of D. magna
toward the plant bed. A preliminary experiment in 80 m2 enclosures in Lake
Stigsholm showed that in the presence of planktivorous fish in high densities
(13-16 m 2) Daphnia spp. sought refuge in the vegetation (Jacobsen et aI., 1997).
Later, more detailed enclosure studies in this lake have shown that addition of 0+
perch (Perea jluviatilis) to fishless enclosures resulted in a differential degree of
habitat shift of zooplankton, dependent on species and development stage (copepods), being particularly high for Daphnia spp. (Jeppesen et aI., in preparation).
DHM can also be an antipredator defense mechanism against invertebrate predators such as Chaoborusjlavieans (Kvam and Kleiven, 1995), involving chemical
cues (Kleiven et aI., 1996). Reverse migration (highest densities in the plant beds
at night) may be observed if plant-associated invertebrate predators such as
odonates are important, as discussed by Lauridsen et al. (this volume, Chapter 13).
DHM induced by invertebrate predators will probably be most important in lakes
with low fish densities.
96
E. Jeppesen et al.
6000~~~~~~~~~~--~----~~~--------.
Ceriodaphnia spp.
•
--0--
Within macrophyte bed
Outside macrophyte bed
4000
T
2000
O+-~====~===o====il===~~--~--~----~~
Bosmina spp.
~ 2000
~
'0
'5
~ 1000
O+-~~~==~===£==~~~~==~--~
Diaphanosoma brachyurum
80
60
40
20
0~~==~~F=~===4~~==~--~
8
11
14
17
20
23
2
5
Time (hour)
Figure 5.2. Diel variations in the abundance of various cladocerans in a 2-mexclosure (open to
small fish and zooplankton) with dense coverage of submerged macrophytes and at a reference
station outside the macrophyte bed in Lake Stigsholm in August. Hatched area shows the dark
period. (Modified from Lauridsen et aI., 1996, by permission of Oxford University Press.)
Macrophytes may also have a repellent effect on various pelagic zooplankters
(Hasler and Jones, 1949; Pennak, 1966; Pennak, 1973; Dorgelo and Heykoop,
1985; Lauridsen and Lodge, 1996). This seems appropriate as the plant beds host
several plant-associated facultative filtrators that most likely are superior competitors to pelagic zooplankton in a plant community in which the production of
periphytic algae is substantially higher than that of phytoplankton (Wetzel, 1975;
Wetzel and S0ndergaard, this volume, Chapter 7). Habitat choice of pelagic
zooplankton therefore seems to be a trade-off between the risk of predation and
optimum food conditions that at least partly seem to be regulated by chemical cues
from either fish or plants.
Diel vertical migration in the pelagic zone (DVM) is well documented and is
most frequently interpreted as being an antipredator defense against predation
5. Fish-Zooplankton Interactions
97
(Ringelberg, 1991; Lampert, 1993) that may involve chemical cues (von Elert and
Loose, 1996). DVM of both pelagic and benthic crustaceans (e.g., Szlaur, 1963;
Whiteside, 1974) may also occur in the plant-rich littoral zone, and various
explanations have been offered. In some studies, it has been interpreted as a
night-time escape from low oxygen concentrations within the macrophyte beds
(Meyers, 1980; Timson and Layboum-Parry, 1985). Others have argued that DVM
in the littoral zone reflects diel variation in predation risk, the predation pressure
being released at night due to reduced visibility and offshore migration by fish
(Jeppesen et aI., in preparation). Further evidence for the importance of fish for
DVM is given by De Stasio (1993), who showed that fish removal in a lake almost
eliminated DVM in the pelagic as well as in the littoral zone. Likewise, Jeppesen
et al. (in preparation) have shown that in the littoral zone especially large-bodied
predation-vulnerable zooplankton such as Daphnia spp. and adult cyclopoid copepods exhibited DVM, whereas smaller forms such as nauplii and rotifers did not.
As expected, DVM was higher in the macrophyte-free area than in the plant bed.
In addition, the degree of DVM increased with increasing density of planktivorous
fish and decreasing plant density. The studies by De Stasio (1993) and Jeppesen et
aI. (in preparation) showed substantial DVM despite the samples integrating the
entire water column except the lower few centimeters. This suggests that the
zooplankton in the littoral zone sought refuge at the sediment surface and in plant
beds perhaps also close to plant surfaces.
DVM in the littoral zone has, however, also been observed in the absence of
fish by Paterson (1993), who argued that DVM of plant-associated forms was
lower in fishless lakes than in fish-rich lakes, which may indicate a lower predation pressure on zooplankton by invertebrate predators than fish. Accordingly,
Paterson (1994) found only marginal differences in the density of cladocerans and
cyclopoid copepods in experiments run at different densities of some important
littoral invertebrate predators (Odonata, Acari, Tanypodinae). Also, Johnson et al.
(1987) and Blois-Heulin et al. (1990) failed to detect a strong impact of large
odonates on cladocerans. By contrast, others have found negative effects of water
mites (Kajak et al., 1968) and Procladius (Dusoge, 1980).
Refuge Effect in Relation to Structural Complexity and Fish Density
Like zooplankton, small fish may seek refuge in the vegetation to avoid predatory
fish and birds (e.g., Carpenter and Lodge, 1986; Gliwicz and Jachner, 1992;
Persson and Crowder, this volume, Chapter 1). The prey fish often prefer sparse
vegetation (Engels, 1988; Phillips et aI., 1996; Jeppesen et aI., 1992, 1997; Stansfield et aI., 1997), which may reflect that the foraging efficiency of fish decreases
with increasing plant density (e.g., Crowder and Cooper, 1982; Savino and Stein,
1982; Anderson, 1984; Diehl, 1988), although there are exceptions to this rule
(Winfield, 1986; Persson and Crowder, this volume, Chapter 1). The study in
Cromes Broad (Phillips et aI., 1996) is an example of aggregation and high predation
on zooplankton by planktivorous fish in sparse vegetation (Fig. 5.3). Accordingly,
98
E. Jeppesen et al.
5
_Roach
DOthers
_Numbers
consumed
4
ill
;:)
n..
if}
u::
90
per point
3
O
J::
120
60
2
....c
'0
n.
0
EC
0
~
E
30
ii5c:
0
v
0
0
Open water
Sparse beds
Dense beds
Figure 5.3. Fish community structure in various habitats in Cromes Broad (1993) shown
by the mean number of fish captured using point sample electrofishing (left column) and
index of point predation (mean ± SE) on zooplankton derived by multiplication of fish
density with the mean number of zooplankton in the fish guts. (Data from Phillips et aI.,
1996.)
zooplankton density was highest in the dense vegetation. Likewise, Lehtovaara
and Sarvala (1984) and Kairesalo and PenttiHi (1990) found higher fish densities
at low density of Equisetum than at high density of this plant, with corresponding
inverse effects on zooplankton densities.
The presence of predatory fish in the vegetation may further complicate the
interaction between planktivorous fish and zooplankton in the littoral zone. On the
one hand, the predation pressure on zooplankton may be reduced as the plankti vorous fish reduce their activity level (Bean and Winfield, 1995; Jacobsen et aI.,
1997; L. Jacobsen, unpublished data) or switch to alternative food sources as a
consequence of a restricted habitat use (Persson, 1993). As an example, Persson
(1993) found that roach in the absence of predators fed on mainly Bosmina sp. but
switched to detritus/algae in the presence of piscivorous perch. This, in tum, led to
an increase in Bosmina density. On the other hand, a high predation risk may drive
the prey fish to move into even the most dense vegetation (Savino and Stein, 1982;
Werner et al., 1983; Persson et aI., 1991; Persson and Ekl6v, 1995) and accordingly a loss of refuge for large-sized zooplankton may occur (Jeppesen et al., in
preparation).
The role of fish density was studied by Schriver et aI. (1995), who conducted
experiments in lOO-m2 enclosures with varying PVI and density of 0+ and 1+
planktivorous fish (three-spined sticklebacks [Gasterosteus aculeatus] and roach
[Rutilus rutilusD in Lake Stigsholm, Denmark. They showed a refuge effect when
PVI was >15-20% if fish density was below ca. 2/m2 (Fig. 5.4). At lower PVI, the
zooplankton were dominated by cyclopoid copepods. At fish densities of 2-4/m2
and PVI at > 15-20%, zooplankton shifted to small cladocerans, and at even higher
99
5. Fish-Zooplankton Interactions
;.
~
"0
01
3-
.... _- .. _---
E
Ul
1800
a.
~
1200
600
0
·.:~:/·······················;;~,·······. ~. .E. .
0""
0
co
+
CII
·c
.£;
a
········;?"o . . yo.
CII
.Ii
0"/0
., 0
6
0
Macrophytes (PVI)
c;,
b
~ .......~ ./~
~
.:·n ...................;;.,/:.......~...cp
<II
~ 10.000
...............~......~ /'B U
rtJ
~
1000
Ol
c:
'N
100
a,
<II
10
~
1
~ . .p
EJ
o
a
. -.. _--.-.
. . .:.7-·.. -··--··.-.--.-.-.-._.. _;./.~_ ...
0
....""/
....
40
Macrophytes (PVI)
Figure 5.4. Biomass of the dominant pelagic cIadocerans (Bosmina + Daphnia; upper
panel) and their potential grazing pressure on phytoplankton (estimated 24-hour ingestion
by Bosmina + Daphnia in % of phytoplankton biomass; lower panel} versus the abundance
of 0+ and I + roach and three-spined sticklebacks (catch per unit effort [CPUE] in traps) and
macrophyte PVI (%) in Lake Stigsholm enclosure experiments involving manipulation of
plants (mainly Potamogeton species) and fish density. (From Schriver et aI., 1995, published with kind permission of Blackwell Science Ltd.)
100
E. Jeppesen et al.
120
-.,100
k..
SO
l
g 60
J
40
..... 20
o.-ro--f---=:::L~
Figure 5.5. Average concentration of various zooplankton in enclosure experiments with
high plant biomass (87-165 g/DW/m2) and contrasting fish densities (0 to 5.5 10-12-cm
perch m 2). The experiment was conducted in hypertrophic Little Mere, England. (Data from
Table 2, Beklioglu and Moss, 1996.)
fish densities, to cyclopoid copepods and rotifers. Thus, submerged macrophytes
acted as refuge for zooplankton against fish predation if PVI was >15-20%, but
the refuge effect almost disappeared-even for small-sized cladocerans-if fish
density exceeded a certain threshold level, in this example ca. 4 fishlm 2. Low
refuge effect and dominance by small zooplankton were also observed in the same
lake in another experiment with a mixed community of 0+ perch and roach (31m2)
and a PVI of 24% (Jeppesen et aI., in preparation). A partial loss of refuge for
large-sized zooplankton within the vegetation has been observed in other studies.
In enclosure experiments with a plant density of 87-165 g dry weight (DW)/m2,
Beklioglu and Moss (1996) found a significant reduction in D. hyalina and
Daphnia cuculata at densities of 2.8 perchlm2 (10-12 cm in length) and a further
reduction of daphnids at a fish density of 5.5/m2 while the densities of small
cladocerans (Ceriodaphnia spp., B. longirostris, Chydorus ovalis) and cyclopoid
copepods increased (Fig. 5.5). No data on total zooplankton biomass are available
from this study, but a marked increase in chlorophyll a as fish densities increased
indicates reduced zooplankton control of phytoplankton. A further evidence of
loss of refuge effect for zooplankton is offered by Persson and Ekl6v (1995), who
used artificial plants with a stem density of 280 m 2 in enclosures and a fish density
of 2.1 m 2 0+ perch and 1+ roach (1: 1). During a 48-day experiment, crustacean
zooplankton were reduced 40-100-fold to densities <1.6 ind/L, with large D.
longispina, as well as large plant-associated cladocerans such as Eurycercus sp.
101
5. Fish-Zooplankton Interactions
Abundance (no 1-')
or
(~g 1-')
o
Cyclopoid cope pods
Chlorophyll a
50
100
150
200
250
300
(~i"
fill) Macrophytes
o Fish+macrophytes
Figure 5.6. Abundance of Bosmina spp., Ceriodaphnia spp., cyclopoid copepods, total
phosphorus, and chlorophyll a in macrophyte-rich enclosures (Elodea canadensis, 216 g
DW m 2 ) without and with fish (3 m2 0+ perch). The results stem from the last sampling date
of the 21-day-long experiment. (Data from Kairesalo et aI., in press.)
reduced to near-zero. In these experiments, no differences were observed whether
plants were present or not. Effects of fish presence were also demonstrated by
Kairesalo et al. (in press), who found that addition of 31m 2 of 0+ perch to enclosures with dense beds (216 g DW/m2) of Elodea canadensis resulted in a major
reduction in the density of Bosmina spp_ (mainly B. longispina) and especially
Ceriodaphnia spp. (mainly C. pulchella) compared with controls without fish, but
densities of cyclopoid copepods did not differ (Fig. 5.6).
These last few examples all show that the refuge effect for Daphnia spp.-and
occasionally also for small-sized cladocerans--even at relatively high plant densities may be partly or totally lost if the density of potentially planktivorous fish
exceeds <2-5/m2 • It is, however, yet to be demonstrated if the same threshold
levels also occur in nature. The enclosure size of these experiments was highly
variable, ranging from 0.7 m2 (Beklioglu and Moss, 1996) to 100 m 2 (Schriver et
ai., 1995), but common to all these studies is that they do not allow discernible die 1
migration of zooplankton and fish between the littoral zone and open water. Lake
Stigsholm experiments indicate that such a migration may reduce the strength of
the interactions between fish and zooplankton in the vegetation (Jeppesen et aI.,
unpublished data). Both roach and perch fry migrate to open water at night, which
in itself may reduce the predation pressure in the littoral zone. This is strengthened
by the fact that, for instance, perch feed especially at dusk and dawn outside the
littoral zone and are less active during the day (Gliwicz and lachner, 1992). Thus,
in practice, the threshold for loss of refuge effect at high plant density may be
higher than indicated by the experiments described above. In sparse beds the fish
threshold for loss of refuge effect may be substantially lower (Schriver et aI., 1995;
Stansfield et al., 1997). The study by Stansfield et al. (1997) suggests, for example,
102
E. Jeppesen et al.
. . . . . . . . . . . . ~. . . . . . /.=.',,~
I
Q)
()
/
c::
CIl
"C
c::
:J
.Q
~"
«
,
,
,/
,/
",I
/
".,
....,
.....
,
......
...
'.\
'
\\
•...
Sep
Figure 5.7. Diagrammatic curves that show abundance of littoral zooplankton during the
summer months in some northern temperate lakes. Curve A, in which there is a midsummer
decline, is the most frequently reported. (From Whiteside, 1988, published with kind
permission of E. Schweizerbartsche Verlagsbuchhandlung, Stuttgart.)
a loss of refuge for Daphnia spp. at a density of 0.25 0+ fish/m 2 • As densities of 0+
fish are often higher than 0.25-5/m2 during the summer (Lehtovaara and Sarvala,
1984; Chick and McIvor, 1994), it is likely that the fish may often have a major
structuring impact on zooplankton in the littoral zone. This is supported by the
studies of the seasonal patterns of microcrustaceans in the littoral zone. Based on
several studies, Whiteside (1988) identified three seasonal patterns (Fig. 5.7): A, a
pre summer peak followed by low densities during July and by a minor peak in
August; B, an almost unimodal pattern with a midsummer maximum; and C, only
an autumn peak. Pattern A is very similar to the one observed in the pelagic zone
of lakes highly influenced by fish predation (Jeppesen et aI., 1997). Whiteside
(1988) claimed that predation is responsible for the summer and autumn declines,
as there were no arguments in favor of deteriorated food conditions. The bimodal
curve (A) is most frequently observed (Whiteside, 1988), indicating that the littoral
refuge effect often tends to disappear during midsummer. Subsequent investigations
by others support Whiteside's suggestions. Yuille (1991) found a major decline in
zooplankton and plant-associated cladocerans in a year with high littoral fish densities
and found minor reductions in the preceding year when fish density was low.
Fish-mediated loss of refuge in the littoral zone does not necessarily mean that
the predation pressure on zooplankton in the total lake per se is high, because
unless the density of older planktivorous fish is high, a predator-mediated aggregation of young fish in the vegetation may lead to a major reduction in the
predation on zooplankton in open water. Thus, Boikova (1986) found a reverse
migration of crustaceans, with daytime densities being 10-100-fold higher in the
pelagic zone than in a dense Elodea canadensis plant bed, whereas the crustaceans
were more evenly distributed at night. The frequently observed avoidance of the
5. Fish-Zooplankton Interactions
103
littoral zone (Hutchinson, 1967; Siebeck, 1969, 1980) also indicates that predation
pressure in the littoral zone is often higher than in the pelagic zone (Gliwicz and
Rybak, 1976; Evans et aI., 1980).
Implications for the Lower Trophic Levels
Due to the aggregation of small fish in the littoral zone and the frequently low
depth, presumably the overall predation pressure on zooplankton in the plant-free
littoral zone may be higher than in the pelagial. Accordingly, the grazing pressure
of zooplankton on phytoplankton, protozoans, and bacterioplankton is likely to be
lower than in the pelagial. Conversely, if plant density is high and fish are not
forced into the vegetation, the cascading effects of grazers may be high due to low
fish predation and the daytime aggregation of pelagic zooplankton in the plant
beds. The spatial differences in grazing pressure of zooplankton between dense
plant beds and open water in the littoral zone are, therefore, expected to be
particularly large. This is supported by a few investigations undertaken so far.
Jeppesen et al. (in preparation) found that in shallow eutrophic Lake Stigsholm,
the daily clearance rates of phytoplankton ranged from very low values of 2% in
the littoral zone outside the plant beds to 3.2% in sparse vegetation (PVI = 24%)
and to values as high as 300% in dense vegetation (PVI = 50%), whereas the
corresponding figures for bacterioplankton clearance were 2.5, 4, and 219%,
respectively (Fig. 5.8). The clearance rates were thus 144-fold (phytoplankton)
and 88-fold (bacterioplankton) higher in the dense beds than in the macrophytefree littoral. Such high clearance rates in dense vegetation had significant cascading effects on the trophic structure within the bed. For example, the densities of
ciliates, phytoplankton, flagellates, and bacterioplankton were 75-, 4.4-, 4-, and
3-fold, respectively, lower inside than outside the beds (Fig. 5.8) (see also
Jeppesen et aI., submitted; Spndergaard and Moss, this volume, Chapter 6;
Spndergaard et aI., this volume, Chapter 15). The significant role played by
zooplankton in the determination of these differences was confirmed by size
fractionation experiments (Jurgens and Jeppesen, this volume, Chapter 16). Earlier
empirical data from experiments involving dense beds of submerged macrophytes
and low fish densities have also shown high zooplankton:phytoplankton biomass
ratios, suggesting a great cascading impact on the lower trophic levels (Irvine et
aI., 1989; Moss et al., 1994; Schriver et aI., 1995; see also Spndergaard and Moss,
this volume, Chapter 6), whereas high fish densities resulted in low ratios inside
and especially outside the littoral vegetation (Schriver et aI., 1995). The high
cascading effect on the lower trophic level within the dense plant beds may not
necessarily reflect the role of aggregating pelagic zooplankton. The beds host
plant-associated filter feeders such as Sida, which may have a potentially high
grazing impact (Stansfield et aI., 1997). The plant beds also host several filterfeeding macroinvertebrates including mussels with high filtering capacity (Ogilvie and Mitchell, 1995). In addition, the plants and epiphytes may help control the
phytoplankton by shading (Straskraba and Pieczynska, 1970) or cause nutrient
E. Jeppesen et al.
104
10000
1000
10000
1000
100
10
1
100
10
1
Rotifers (no. 1-1)
Cyclopoid cope pods (no. 1-1)
Small cladocerans (no. 1-1)
Daphnia (no. 1- 1)
Bacterioplankton .9razing ~mll-l d- 1)
Phytoplankton grazing (mrl- d-1)
10000
1000
10000
1000
100
10
1
100
10
1
Ciliates (no. I-l x 10-2)
Flagellates (no. 1-1x 10-3 )
Bacterioplankton (no. ml- 1 x 10-6 )
Chlorophyll a (~g 1-1)
Figure S.S. Abundance (no. L-I) of various zooplankton and microbial components,
chlorophyll a (J.Lg L-1), and zooplankton (>140 /.lm) clearance (mIlL/day) of phytoplankton
and bacterioplankton in Lake Stigsholm enclosures with contrasting densities of submerged
macrophytes (HM, plant volume infested (PVI) = 50%; LM, PVI = 24%; M-, without
macrophytes). Note the logarithmic scale. (Data from Jeppesen et al., submitted.)
5. Fish-Zooplankton Interactions
105
limitation (Kairesalo et al., in press). Multiple factors may thus contribute to the
relatively high zooplankton:phytoplankton biomass ratio and the stronger topdown control of phytoplankton in the dense plant beds with low fish densities.
Although aggregation of zooplankton in the vegetation has a substantial impact
on both the trophic structure and phytoplankton biomass within the plant bed, we
do not know the extent to which night-time migration to the pelagic zone will
influence open water phytoplankton. Lauridsen et al. (1996) estimated that a 3%
coverage of the lake surface area with 2-m diameter patches of dense Potamogeton pectinatus beds is sufficient for a night-time doubling of the density of
Ceriodaphnia spp. and B. longirostris in Lake Stigsholm with a low natural plant
coverage, and this must be assumed to have a great impact on the phytoplankton
grazing pressure. It does, however, require that the zooplankton are widely spread
throughout the pelagic zone during night. There are still no measurements of how
far zooplankton migrate horizontally. DVM studies have shown night-time migration of >30 m (e.g., Geller et al., 1992) and a mean migration velocity of 0.2 cm/
sec for large-sized Daphnia species and at maximum speed as much as .52 cm/sec (S.1. Dodson, personal communication). If these figures can be transferred to DHM, the zooplankton may be able to exploit the entire pelagic zone
in many shallow lakes. An additional consideration is that plants in shallow
lakes are often not restricted to the nearshore area but can be found in patches
in large parts or in somewhat deeper areas of the lake, further increasing the
possibilities of night-time exploitation of the pelagic zone. We, therefore,
predict that in many shallow lakes the possibility of seeking refuge in the
vegetation during the day increases grazer control of the phytoplankton in
open water and thus contributes to maintaining those lakes with comprehensive and dense macrophyte coverage in the clearwater state. The establishment
of macrophyte refuges protected from waterfowl grazing has been proposed as
a restoration measure to complement nutrient-loading reductions in shallow
lakes (Moss, 1990; Jeppesen et aI., 1991). Establishment of numerous small and
dense refuges should therefore result in much higher densities of migrating
cladocerans than a few large refuges. The higher density of cladocerans will
ensure a greater filtering capacity within the beds during the day and in open water
during the night. Per unit area, small and dense macrophyte refuges may be better
able to promote a shift to a clearwater stage than larger ones with low macrophyte
density (Lauridsen et aI., 1996; Jeppesen et al., 1997).
Changes in Interactions Along a Nutrient Gradient
Changes in nutrient levels affect both the abundance and composition of macrophytes (Wetzel, 1975) and fish (Persson et aI., 1991; Jeppesen et aI., 1997), which
will alter the zooplankton refuge efficiency of the plants against fish predation.
Empirical data are scarce, but we propose various hypotheses that may help
initiate future discussions and tests. We restrict ourselves to northern European
lakes.
106
E. Jeppesen et al.
With increasing nutrient levels, the depth limit of submerged macrophytes
decreases (Chambers and Kalff, 1985), but at the same time the biomass per unit
of area and stem density of plants increase (Wetzel, 1975). This supposedly leads
to a reduction of the total refuge area for zooplankton, but in the remaining
plant-filled areas the refuge effect will increase. Whether increased density compensates for lower area coverage is open to debate, but with the step-like increases
observed by Schriver et ai. (1995) in the refuge effect at high plant density, it may
be presumed that this is indeed the case. At the highest nutrient levels, submerged
macrophytes most often totally disappear due to light limitation caused by phytoplankton (but see Moss et aI., 1997), leaving only floating-leaved plants and reed
belts, which often have a comparatively low refuge effect (Gliwicz and Rybak,
1976; Winfield, 1986; Venugopal and Winfield, 1993) due to low stem density.
Presumably, therefore, the refuge effect at a fixed density of planktivorous fish is
greatest in slightly eutrophic lakes in which plant density is high and the area
covered not yet severely reduced.
Simultaneously with the changes in density and distribution of plants, there are
also plant composition changes. In northern temperate lakes, the succession is
often from characeans to elodeids in hard water lakes and from isotids to elodeids
in softwater lakes. Isoetids are small dense rosette plants that may act as an
efficient refuge among the leaves. However, due to small stature, their general
effect as a refuge is probably poor. Characeans often form dense beds with a high
areal biomass. Potentially, they may therefore act as an efficient refuge against
predation by fish: Diehl (1988) has shown that high density of characeans results
in high density of macroinvertebrates. We do not, however, know if this is true for
zooplankton. Elodeids may also appear in high densities, but the biomass per unit
volume is often smaller than that of characeans (Diehl, 1988), indicating that
characeans potentially may act as a better refuge than elodeids.
The picture is further complicated by changes in the density and the relative
contribution of some fish species along the nutrient gradient in lakes (Persson et
al., 1988; Jeppesen et aI., 1990). In eutrophic and hypereutrophic lakes, planktibenthivorous fish such as roach and bream dominate, being the most efficient
fouragers in the pelagic habitat. The preference for the pelagial may be further
strengthened by the fact that plant-associated pike (Esox lucius L.) is often the
dominant piscivore in such lakes (Grimm and Backx, 1990), making it less
favorable for prey fish to forage in the vegetation. The pattern will be different if
the dominant species is the pelagic forager zander (Stizostedion lucioperca), but
this species is often not abundant in lakes with extensive growth of submerged
macrophytes. Thus, the efficiency of plant beds as a refuge for zooplankton will
often be high in eutrophic lakes. The aggregation of zooplankton is further
strengthened by the high risk of predation in the pelagic zone.
In less eutrophic to mesotrophic lakes, the impact of predatory fish increases,
perch become more important, and the foraging conditions for predatory fish
improve due to, for instance, increased transparency. Prey fish thus seek refuge in
the vegetation. As planktivorous fish density is relatively high, it is expected that
aggregation of young planktivorous fish in the vegetation will be particularly high
5. Fish-Zooplankton Interactions
107
D. Mesotrophic
A. Hypertrophic
-- , ,
. . . --.. . ,,' A
,
".
/ ___ ."..~
B. Eutrophic -
high macrophyte density
, ... -, ,
,,
<J)
~
E
o
I
CO
I
\
E. Oligotrophic
\
\
I
I
\
\
\
-- C. Eutrophic -
' . . ":.t
.
..
...
......
low macrophyte density
November
April
Seasonal cycle
Figure 5.9. Conceptual model showing how the seasonal dynamics of microcrustacean
biomass (solid line) in the littoral zone is expected to vary under different nutrient conditions. The broken line represents the average biomass of submerged macrophytes in plantcovered areas.
and the refuge effect correspondingly low. In mesotrophic-oligotrophic lakes, the
predator control of planktivorous fish increases. The aggregation is expected to
remain high due to a larger share of predatory fish, but the lower abundance of
planktivores and lower fish density counteract the effect. The question is if the
refuge effect at a given plant density will be higher or lower than in slightly
eutrophic lakes. In oligotrophic lakes, the refuge effect for prey fish due to low
plant density and plant height will probably be poor, so that the refuge effect for
zooplankton may increase.
It is difficult to combine these multiple and complex interactions into a common conceptual model, but we tentatively predict (Fig. 5.9) that
• In hypertrophic lakes without submerged macrophytes, the refuge effect for
zooplankton in the littoral zone dominated by reeds is poor throughout the
108
•
•
•
•
E. Jeppesen et al.
summer. The density of zooplankton and to a lesser extent plant-associated
crustaceans is low and dominated by small forms (Fig 5.9A).
In eutrophic lakes with high areal coverage of submerged macrophytes and
high PVI, the refuge effect will be high throughout the summer period,
during which planktivorous fish stay in the pelagic zone or in the sparse
vegetation. The density of macrocrustaceans in the plant bed will be high,
apart from a period during early summer before macrophyte density has
become high. (Fig. 5.9B).
In eutrophic lakes with low PVI, the pattern approaches the one suggested for
hypereutrophic lakes (Fig. 5.9C).
In meso-slightly eutrophic lakes, 0+ fish seek refuge in the vegetation during
mid-summer, and Whiteside's type A (bimodal) or C (unimodal with autumn
maximum) response of microcrustacean density (see Fig. 5.7) can be observed, depending on whether the abundance ofplanktivor.ous fish (:2 I year)
is low or high. Very high plant density (e.g., of characeans) may result in a
response resembling that of eutrophic lakes. (Fig. 5.9D).
In oligotrophic lakes, the refuge effect will often be low due to low plant
height (Fig. 5.9E).
The patterns described cannot be applied to brackish lakes in northern Europe that
deviate substantially in trophic structure and dynamics from freshwater lakes
(Leah et aI., 1978; Jeppesen et aI., 1994, this volume, Chapter 31). Several factors
indicate that the refuge effect for zooplankton in nutrient-rich brackish lakes is
poor due to aggregation of both sticklebacks and Neomysis in the plant beds
(Jeppesen et aI., this volume, Chapter 31).
Future Research Needs
Although much new information has appeared during recent years about the
impact of macrophytes on fish-zooplankton interactions, there are still several
unclarified questions. We know that DHM does occur in lakes, but little is known
about the distances covered by the zooplankton. This is interesting from a theoretical point of view, but it also has practical implications. As mentioned, the use of
macrophyte implantations has been suggested as a restoration tool in lakes. To
ensure optimum placement of these macrophyte enclosures to obtain the highest
night-time grazing effect of zooplankton in the pelagic zone, more information on
the potential migration distances is needed. Also, we know little about how the
zooplankton find their way back from the pelagic to the plant-covered areas during
the daytime. Moreover, it is unknown what the migration pattern would look like
if zooplankton are influenced by both fish and invertebrate predators inhabiting
spatially segregated areas (e.g., fish in the pelagic zone and odonates, water mites,
etc., in the littoral zone). In addition, cost-benefit analyses are required. When will
horizontal migration be cost-efficient and how do such estimates alter along a
gradient in food supply and predation risk in the vegetation and the pelagic zone?
5. Fish-Zooplankton Interactions
109
What influence does the size of the plant-filled areas have? To answer these questions,
we suggest both intensive laboratory and field studies, including detailed studies of
zooplankton population dynamics and the use of a modeling approach.
Most of the former studies of the interactions between fish, macrophytes, and
zooplankton were undertaken with only one or two species of prey fish and
typically one species of predatory fish. Because interactions between a given fish
species and zooplankton may change in the presence of other species (Persson and
EklOv, 1995), there is a great need for multispecies experiments. A better insight
into the interactions between fish and zooplankton also requires a more thorough
knowledge of the feeding behavior of the various zooplanktivorous fish. This is
crucial because it is not possible from data on distribution alone to determine
where and when the interactions between fish and zooplankton are particularly
strong.
It is necessary to undertake investigations under more natural conditions than
hitherto has been the case. This means on a scale that allows horizontal migration for
both zooplankton and fish (i.e., a large scale and whole-lake basis). Empirical studies
of seasonal and diel variations in fish and zooplankton in the pelagic and littoral
zones, with contrasting nutrient levels, fish communities, and macrophyte abundance and composition, may contribute to the understanding of natural interactions.
Also, the inclusion of quantitative paleoecological investigations, including reconstruction of fish and submerged macrophytes (Jeppesen et aI., 1996), will add
to our understanding, and it may increase our knowledge about long-time perturbations, factors that are poorly elucidated from the existing short monitoring series
and short-term enclosure and whole-lake experiments (Anderson, 1995). We need
to know more about the role of plant-associated microcrustaceans and their interactions with zooplankton staying temporarily or permanently in the open water
among the plants. Stable isotope (l3C and 15N) analyses and grazing measurements on
radio-labeled periphyton may be useful methods in such studies. Finally, there is a
great need for studies of how density and composition of submerged macrophytes
affect the interaction between fish, Neomysis, and zooplankton in brackish lakes.
Acknowledgments. The assistance of the technical staff of the National Environmental Research Institute, Silkeborg, is gratefully acknowledged. Technical assistance was provided by K. M0gelvang and A.M. Poulsen. The study was supported
by the Centre for Freshwater Environmental Research. We thank Ramesh Gulati,
John Iwan Jones, Brian Moss, Greg Cronin, and Martin S0ndergaard for valuable
comments on the manuscript.
References
Anderson, N.J. Temporal scale, phytoplankton ecology and palaeoecology. Freshwat. BioI.
34:367-378; 1995.
Anderson, O. Optimal foraging by largemouth bass in structured environments. Ecology
65:851-861; 1984.
Bean, C.W.; Winfield, 1.1. Habitat use and activity patterns of roach (Rutilus rutilus), rudd
(Scardinius erythrophtalmus (L.», perch (Percafluviatilis) and pike (Esox lucius) in the
110
E. Jeppesen et al.
laboratory: the role of predation threat and structural complexity. Ecol. Freshwat Fish
4:37-46; 1995.
Beklioglu, M.; Moss, B. Mesocosm experiments on the interaction of sediment influence,
fish predation and aquatic plants with the structure of phytoplankton and zooplankton
communities. Freshwat. BioI. 36:315-325; 1996.
Benndorf, J. Food web manipulation without nutrient control: a useful strategy in lake
restoration? Schweiz. Z. Hydro!. 49:237-248; 1987.
Blois-Heulin, C.; Crowley, P.H.; Arrington, M.; Johnson, D.M. Direct and indirect effects of
predators on the dominant invertebrates of two freshwater littoral communities.
Oecologia 84:295-306; 1990.
Boikova, O.S. Horizontal distribution of crustaceans in Lake Glubokoe. Hydrobiologia
141: 113-123; 1986.
Brooks, J.L.; Dodson, S.l. Predation, body size and composition of plankton. Science
150:28-35; 1965.
Carpenter, S.R.; Kitchell, J.P., eds. The trophic cascade in lakes. Cambridge: Cambridge
University Press; 1993.
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Carpenter, S.R.; Kitchell, J.E; Hodgson, J.R. Cascading trophic interactions and lake
productivity. BioScience 35:635-639; 1985.
Chambers, P.A; Kalff, J. Depth distribution and biomass of submersed aquatic macrophyte
communities in relation to Secchi depth. Can. J. Fish. Aquat. Sci. 47: 1929-1936; 1985.
Chick, J.H.; Mel vor, c.c. Patterns in the abundance and composition of fishes among beds
of different macrophytes: viewing a littoral zone as a landscape. Can. J. Fish. Aquat. Sci.
51 :2873-2882; 1994.
Christoffersen, K.; Riemann, B.; Klysner, A; S¢ndergaard, M. Potential role of
zooplankton in structuring a plant community in eutrophic lake water. Limno!.
Oceanogr. 38:561-573; 1993.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegill and their prey. Ecology 63: 1802-1813; 1982.
Cryer, M.; Townsend, c.R. Spatial distribution of zooplankton in a shallow eutrophic lake,
with a discussion of its relation to fish predation. J. Plankton Res. 10:487-501; 1988.
Cryer, M.; Pierson, G.; Townsend, c.R. Reciprocal interactions between roach Rutilus
rutilus, and zooplankton in a small lake: prey dynamics and fish growth and recruitment.
Limnol. Oceanogr. 31: 1022-1038; 1986.
Cyr, H.; Downing, J.A Empirical relationships of phytomacrofaunal abundance to plant
biomass and macrophyte bed characteristics. Can. J. Fish. Aquat. Sci. 45:976-984;
1988.
Davies, J. Evidence for a diurnal horizontal migration in Daphnia hyalina lacustris Sars.
Hydrobiologia 120:103-105; 1985.
De Bernardi, R.; Guisanni, G. Are blue-green algae suitable food for zooplankton? An
overview. Hydrobiologia 200/201 :29-41; 1990.
De Meester, L.; Maas, S.; Dierckens, K.; Dumont, H.J. Habitat selection of patchiness in
Scapholeberis: horizontal distribution and migration of S. mucronata in a small pond. J.
Plankton Res. 15:1129-1139; 1993.
DeMott, W.R. The role of taste in food selection by freshwater zooplankton. Oecologia
69:224-240; 1986.
De Stasio, B.T., Jr. Diel vertical and horizontal migration by zooplankton: population
budgets and the diurnal deficit. Bull. Mar. Sci. 53:44--64; 1993.
Diehl, S. Foraging efficiency of three freshwater fishes: effects of structural complexity and
li,ght. Oikos 53:207-214; 1988.
Dorgelo, J.; Heykoop, M. Avoidance of macrophytes by Daphnia longispina. Verh. Int.
Verein. Limnol. 22:3369-3372; 1985.
5. Fish-Zooplankton Interactions
11l
Downing, J.A; Plante, C.; Lalonde, S. Fish production correlated with primary productivity
and the morphoedaphic index. Can. J. Fish. Aquat. Sci. 47:1929-1936; 1990.
Dusoge, K. The occurrence and role of the predatory larvae of Procladius Skuse (Chironomidae, Diptera) in the benthos of Lake Sniardy. Eko1. Pol. 28:155-186; 1980.
Engels, S. The role and interactions of submersed macrophytes in a shallow Wisconsin lake.
J. Freshwat. Eco1. 4:329-340; 1988.
Evans, M.S.; Bethany, E.H.; Sell, D.W. Seasonal features of zooplankton assemblages in
the nearshore area of southeastern Lake Michigan. J. Great Lakes Res. 6:275-289; 1980.
Geller, W.; Pinto-Coelho, R; Pauli, H.-R. The vertical distribution of zooplankton (Crustacea, Rotatoria, Ciliata) and their grazing over the diurnal and seasonal cycles in Lake
Constance. Arch. Hydrobio1. Beih. Ergebn. Limno1. 35:79-85; 1992.
Gliwicz, Z.M.; Jachner, A Diel migration of juvenile fish: a ghost of predation past or
present. Arch. Hydrobio1. 124:385-410; 1992.
Gliwicz, Z.M.; Rybak, J.1. Zooplankton. In: Pieczynska, E., ed. Selected problems of lake
littoral ecology. Warsaw: University of Warsaw; 1976:69-96.
Grimm, M.P.; Backx, J.J.G.M. The restoration of shallow eutrophic lakes, and the role of
northern pike, aquatic vegetation and nutrient concentration. Hydrobiologia 2001201:
557-566; 1990.
Gulati, RD.; Lammens, E.H.R.R; Meijer, M.-L.; van Donk, E. Biomanipulation tool for
water management. Hydrobiologia 2001201:1-628; 1990.
Hanson, J.R.; Leggett, W.e. Empirical prediction offish biomass and weight. Can. J. Fish.
Aquat. Sci. 39:257-263; 1982.
Hasler, AD.; Jones, E. Demonstration of the antagonistic action of large aquatic plants on
algae and rotifers. Ecology 30:359-364; 1949.
He, X.; Wright, R An experimental study ofpiscivore-planktivore interactions: population
and community responses to predation. Can. J. Fish. Aquat. Sci. 49:1176-1185; 1992.
Hrbacek, J.; Dvorakova, M.; Korinek, M.; Prochazkova, L. Demonstration of the effect of
the fish stock on the species composition of zooplankton and the intensity of metabolism
ofthe whole plankton association. Verh. Int. Verein. Limno1. 14: 192-195; 1961.
Hutchinson, G.E. A treatise on limnology II. New York: J. Wiley & Sons; 1967.
Irvine, K.; Moss, B.; Balls, H. The loss of submerged plants with eutrophication II.
Relationships between fish and zooplankton in a set of experimental ponds, and conclusions. Freshwat. BioI. 22:89-107; 1989.
Jacobsen, L.; Perrow, M.R; Landkildehus, E; Hj~rne, M.; Lauridsen, T.L.; Berg, S. Interactions between piscivores, zooplanktivores and zooplankton in submerged macrophytes: preliminary observations from enclosure and pond experiments. Hydrobiologia
3421343:197-205; 1997.
Jakobsen, P.J.; Johnsen, G.H. The influence of predation on horizontal distribution of
zooplankton species. Freshwat. BioI. 17:501-507; 1987.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; S~ndergaard, M.; Mortensen, E.; Sortkja:r, 0.;
Olrik, K. Fish manipulation as a lake restoration tool in shallow, eutrophic, temperate
lakes. 2: Threshold levels, long-term stability and conclusions. Hydrobiologia 200/201:
219-227; 1990.
Jeppesen, E.; Kristensen, P.; Jensen, J.P.; S~ndergaard, M.; Mortensen, E.; Lauridsen, T.L.
Recovery resilience following a reduction in external phosphorus loading of shallow,
eutrophic Danish lakes: duration, regulating factors and methods for overcoming resilience. Mem. 1st. lta1. Idrobiol. 48:127-148; 1991.
Jeppesen, E.; Sortkja:r, 0.; S~ndergaard, M.; Erlandsen, M. Impact of a trophic cascade on
heterotrophic bacterioplankton production and biomass in two shallow fish-manipulated
lakes. Arch. Hydrobio1. Ergebn. Limnol. 37:219-231; 1992.
Jeppesen, E.; S~ndergaard, M.; Kanstrup, E.; Petersen, B.; Henriksen, RB.; Hammersh~j,
M.; Mortensen, E.; Jensen, J.P.; Have, A. Does the impact of nutrients on the biological
structure and function of brackish and freshwater lakes differ? Hydrobiologia 275/276:
15-30; 1994.
112
E. Jeppesen et al.
Jeppesen, E.; Agerbo Madsen, E.; Jensen, J.P.; Anderson, N.J. Reconstructing the past
density of planktivorous fish and trophic structure from sedimentary zooplankton fossils: a surface sediment calibration data set from shallow lakes. Freshwat. BioI. 36: 115127; 1996.
Jeppesen, E.; lensen, J.P.; S0ndergaard, M.; Lauridsen, T.; Pedersen, LJ.; Jensen, L.
Top-down control in freshwater lakes: the role of nutrient state, submerged macrophytes
and water depth. Hydrobiologia 342/343: 151-164; 1997.
lohnson, D.M.; Pierce, C.L.; Martin, T.H.; Watson, e.N.; Bohanan, R.E.; Crowley, P.H.
Prey depletion by odonate larvae: combining evidence from multiple field experiments.
Ecology 68:1459-1465; 1987.
Jiirgens, K. Impact of Daphnia on planktonic microbial food webs-a review. Microb.
Food Webs 8:295-324; 1994.
Kairesalo, T. Diurnal fluctuations within a littoral plankton community in oligotrophic Lake
Paajarvi, southern Finland. Freshwat. BioI. 10:533-537; 1980.
Kairesalo, T.; Penttila, S. Effect of light and water flow on the spatial distribution of littoral
Bosmina longispina Leydig (Cladocera). Verh. Int. Verein. Limnol. 24:682-687; 1990.
Kairesalo, T.; Tatrai, 1.; Luokkanen, E. Impacts of waterweed (Elodea canadensis Michx)
on plankton-fish interactions in lake littoral. Verh. Int. Verein. Limnol. (in press).
Kajak, Z.; Dusoge, K.; Stanczykowska, A. Influence of mutual relations of organisms,
especially Chironomidae, in natural benthic communities, on their abundance. Ann.
Zool. Fenn. 5:49-56; 1968.
Kleiven, O.T.; Larssen, P.; Hob<ek, A. Direct distributional response in Daphnia pulex to a
predator kairomone. J. Plankton Res. 18:1341-1348; 1996.
Kvam, O.Y.; Kleiven, O.T. Diel horizontal migration and swann formation in Daphnia in
response to Chaoborus. Hydrobiologia 307: 177-184; 1995.
Lampert, W. Ultimate causes of diel vertical migration of zooplankton: new evidence for
the predator-avoidance hypothesis. Arch. Hydrobiol. Beih. Ergebn. Limnol. 39:79-88;
1993.
Lauridsen, T.L.; Buenk, 1. Diel changes in the horizontal distribution of zooplankton in the
littoral zone of two shallow eutrophic lakes. Arch. Hydrobiol. 137: 161-176; 1996.
Lauridsen, T.L.; Lodge, D. Avoidance by Daphnia magna fish and macrophytes: chemical
cues and predator-mediated use of macrophyte habitat. Limnol. Oceanogr. 41 :794-798;
1996.
Lauridsen, T.L.; Pedersen, L.J.; Jeppesen, E.; S0ndergaard, M. The importance of macrophyte bed size for cladoceran composition and horizontal migration in a shallow lake. J.
Plankton Res. 18:2283-2294; 1996.
Leah, R.T.; Moss, B.; Forrest, D.E. Experiments with large snails in a fertile, shallow,
brackish lake. Hickling Broad, United Kingdom. Int. Rev. Ges. Hydrobiol. 63:291-310;
1978.
Lehtovaara, A.; Sarvala, J. Seasonal dynamics oftotal biomass and species composition of
zooplankton in the littoral of an oligotrophic lake. Verh. Int. Verein. Limnol. 22:805810; 1984.
Leibold, M.A. Resource edibility and the effects of predators and productivity on the
outcome of trophic interactions. Am. Nat. 134:922-949; 1990.
Luecke, e.; Vanni, M.J.; Magnuson, 1.1.; Kitchell, J.F.; Jacobson, P.I. Seasonal regulation of
Daphnia populations by planktivorous fish: implications for the clearwater phase. Limnol.
Oceanogr. 35:1718-1733,1990.
McQueen, D.I.; Post, 1.R.; Mills, E.L. Trophic relationships in freshwater pelagic ecosystems. Can. J. Fish. Aquat. Sci. 43:1571-1581; 1986.
Meyers, D.G. Diurnal vertical migration in aquatic microcrustaceans: light and oxygen
responses of littoral zooplankton. In: Kerfoot, W.e., ed. Evolution and ecology of
zooplankton communities. Hanover, NH: University Press of New England; 1980:
80-90.
5. Fish-Zooplankton Interactions
113
Mills, E.L.; Forney, lL.; Wagner, KJ. Fish predation and its cascading effect on the Oneida
Lake food chain. In: Kerfoot, W.e.; Sih, A., eds. Predation: direct and indirect effects on
aquatic communities. Hanover, NH: University Press of New England; 1987:118-131.
Moss, B. Engineering and biological approaches to the restoration from eutrophication of
shallow lakes in which aquatic plant communities are important components. Hydrobiologia 200/201:367-378; 1990.
Moss, B.; McGowan, S.; Carvalho, L. Determination of phytoplankton crops by top-down
and bottom-up mechanisms in a group of English lakes, the West Midland Meres.
Limnol. Oceanogr. 39: 1020-1029; 1994.
Moss, B.; Beklioglu, M.; Carvalho, L.; Kilinc, S.; McGowan, S.; Stephen, D. Vertically
challenged limnology: contrasts between deep and shallow lakes . Hydrobiologia 3421
343:257-267; 1997.
Ogilvie, S.H.; Mitchell, S.F. A model of mussel filtration in a shallow New Zealand lake,
with reference to eutrophication control. Arch. Hydrobiol. 133:471-482; 1995.
Paterson, M. The distribution of microcrustacea in the littoral zone of a freshwater lake.
Hydrobiologia 263: 173-183; 1993.
Paterson, M. Invertebrate predation and seasonal dynamics of mic:rocrustaceans in the
littoral zone of a fishless lake. Arch. Hydrobiol. 99: 1-36; 1994.
Pennak, RW. Structure of zooplankton populations in the littoral macrophyte zone of some
Colorado lakes. Trans. Am. Microsc. Soc. 85:329-349; 1966.
Pennak, RW. Some evidence for aquatic macrophytes as repellents for a limnetic species of
Daphnia. Int. Rev. Ges. Hydrobiol. 58:569-576; 1973.
Persson, L. Predator-mediated competition in prey refuges: the importance of habitat
dependent prey resources. Oikos 68: 12-22; 1993.
Persson, L.; Eklov, P. Prey refuges affecting interactions between piscivorous perch and
juvenile perch and roach. Ecology 76:763-784; 1995.
Persson, L.; Anderson, G.; Hamrin, S.F.; Johansson, L. Predation regulation and primary
production along the productivity gradient of temperature lake ecosystems. In: Carpenter, S.R., ed. Complex interactions in lake communities. New York: Springer-Verlag;
1988:45-65.
Persson, L.; Diehl, S.; Johanson, L.; Andersson, G.; Hamrin, S.F. Shifts in fish communities
along the productivity gradient of temperate lake-patterns and the importance of sizestructured populations. 1. Fish BioI. 38:281-293; 1991.
Phillips, G.L.; Perrow, M.R; Stansfield, 1. Manipulating the fish-zooplankton interaction
in shallow lakes: a tool for restoration. In: Greenstreet, S.P.R; Tasker, M.L., eds. Aquatic
predators and their prey. Oxford: Blackwell Scientific Publications; 1996: 174-183.
Ringelberg, J. Enhancement of the phototactic reaction in Daphnia hyalina by a chemical
mediated by juvenile perch (Percafluviatilis). J. Plankton Res. 13:17-25; 1991.
Samelle, O. Nutrient enrichment and grazer effects of phytoplankton in lakes. Ecology
73:551-560; 1992.
Savino, J.; Stein, R.A. Predator-prey interaction between largemouth bass and bluegills as
influenced by simulated, submersed vegetation. Trans. Am. Fish. Soc. 111:255-266;
1982.
Schriver, P.; B¢gestrand, J.; Jeppesen, E.; S¢ndergaard, M. Impact of submerged macrophytes on the interactions between fish, zooplankton and phytoplankton: large-scale
enclosure experiments in a shallow lake. Freshwat. BioI. 33:255-270; 1995.
Shapiro, J.; Lamarra, Y.; Lynch, M. Biomanipulation: an ecosystem approach to lake
restoration. In: Brezonik, P.L.; Fox, J.F., eds. Water quality management through biological control. Gainesville, FL: University of Florida; 1975:85-96.
Siebeck, O. Spatial orientation of planktonic crustaceans. 1. The swimming behaviour on a
horizontal plane. Verh. Int. Verein. Limnol. 17:831-840; 1969.
Siebeck, O. Optical orientation of pelagic crustaceans and its consequence in the pelagic
and the littoral zones. In: Kerfoot, W.e., ed. Evolution and ecology of zooplankton
communities. Hanover, NH: University Press of New England; 1980:28-38.
114
E. Jeppesen et al.
S!Ilndergaard, M.; Jeppesen, E.; Berg, S. Pike (Esox lucius L.) stocking as a biomanipulation
tool. 2. Effects on lower trophic levels in lake Lyng (Denmark). Hydrobiologia 3421
343:319-325; 1997.
Stansfield, J.H.; Perrow, M.R.; Tench, L.D.; Jowitt, A.J.D.; Taylor, A.A.L. Submerged
macrophytes as refuges for grazing Cladocera against fish predation: observations on
seasonal changes in relation to macrophyte cover and predation pressure. Hydrobiologia
342/343:229-240; 1997.
Straskraba, M.; Pieczynska, E. Field experiments on shading effect by emergents on littoral
phytoplankton and periphyton production. Rozpravy Ceskoslovenske Akademie Ved.
Iada Matematickych a Pfirodnich Ved. 80:7-30; 1970.
.
Szlauer, L. Diurnal migrations of minute invertebrates inhabiting the zone of submerged
hydrophytes in a lake. Schweiz. A. Hydro]. 25:56-64; 1963.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Timson, S.; Laybourn-Parry, J. The behavioural responses and tolerance of freshwater
benthic cyc1opoid copepods to hypoxia and anoxia. Hydrobiologia 127:257-263; 1985.
Venugopal, M.N.; Winfield, I.J. The distribution of juvenile fishes in a hypereutrophic
pond: can macrophytes potentially offer a refuge for zooplankton? J. Freshwat. Ecol.
8:389-396; 1993.
von Elert, E.; Loose, C.J. Predator-induced diel vertical migration in Daphnia: enrichment
and preliminary chemical characterization of a kairomone exuded by fish. J. Chern.
Ecol. 22:885-895; 1996.
Yuille, T. Abundance, standing crop and production of microcrustacean populations (Cladocera, Copepoda) in the littoral zone of Lake Biel, Switzerland. Arch. Hydrobiol.
123:165-185; 1991.
Yuille, T.; Maurer, V. Body mass of crustacean zooplankton in Lake Biel: a comparison
between pelagic and littoral communities. Verh. Int. Verein. Limnol. 24:938-942; 1991.
Walls, M.; Rajasilta, M.; Sarvala, J.; Salo, J. Diel changes in horizontal microdistribution of
littoral c1adocera. Limnologica 20:253-258; 1990.
Werner, E.E.; Oilliam, J.F.; Hall, D.J.; Mittelbach, 0.0. An experimental test of the effects
of predation risk on habitat use in fish. Ecology 64:1540-1548; 1983.
Wetzel, R.O. Limnology. Philadelphia: W.B. Saunders; 1975.
Whiteside, M.e. Chydorid (Cladocera) ecology: seasonal patterns and abundance ofpopulations in Elk Lake, Minnesota. Ecology 55:538-550; 1974.
Whiteside, M.C. 0+ fish as major factors affecting abundance patterns of littoral zooplankton. Verh. Int. Verein. Limnol. 23:1710-1714; 1988.
Winfield, I.J. The influence of simulated aquatic macrophytes on the zooplankton consumption rate of juvenile roach, Rutilus rutilus, rudd, Scardinius erythrophthalmus, and
perch, Percafluviatilis. J. Fish BioI. 29:37-48; 1986.
6. Impact of Submerged Macrophytes on
Phytoplankton in Shallow Freshwater Lakes
Martin Sj2jndergaard and Brian Moss
Introduction
The ubiquity of phytoplankton, and its fundamental importance as a primary
producer and mediator of many major biological processes in lakes, has led to
comprehensive research on its biology. Its importance for water quality and its
increasing predominance as the main primary producer at the expense of submerged macrophytes in shallow lakes follow from increased nutrient loading.
Often, macrophytes have completely disappeared, and nutrients are so abundant
that it is difficult to conceive much bottom-up control of phytoplankton through
especially phosphorus. Restriction of the nutrient loading to reduce the amount of
phytoplankton and to increase water clarity and restore a more diverse biological
structure has now started to reverse this process in many areas. Simultaneously, we
can anticipate a renewed importance of submerged macrophytes in many lakes.
Most research has, however, been on phytoplankton in the pelagic environment, and few studies have focused on phytoplankton in relation to the presence of
submerged macrophytes. Thus, the existence and relative importance of top-down
control of phytoplankton by zooplankton and/or bottom-up control through nutrients, which is well known from the pelagic, is poorly documented from macrophyte beds. It is largely unknown whether the mechanisms in the macrophyte beds
are similar or to what extent different densities of macrophytes may influence a
trophic cascade. By their structuring effect, macrophytes create an environment
that is fundamentally different from that of the open water and that potentially may
115
116
M. S0ndergaard and B. Moss
have great impact on the interactions between the different trophic levels (Jeppesen et al., this volume, Chapter 5).
In this chapter, we describe how submerged macrophytes may affect phytoplankton biomass and community structure. We focus on temperate shallow lakes
with the potential for an extensive macrophyte cover. It is not our intention to give
a comprehensive review but to highlight some of the clearest and, in our opinion,
most important relationships. We supplement these with some of our own experience and results, particularly from large-scale enclosure experiments in the
shallow and eutrophic Lake Stigsholm, Denmark, and Little Mere in the United
Kingdom, where macrophyte and fish densities have been manipulated. We also
try to create a framework that defines hypotheses about which phytoplankton
biomass and species differences to expect along nutrient, zooplankton, and macrophyte gradients.
Effect of Macrophytes on Phytoplankton Biomass
Assuming a given external flux of nutrients and fish-mediated grazing pressure
from zooplankton, several mechanisms may be responsible for the general observation that a lower phytoplankton biomass is found in the presence compared with
the absence of macrophytes. The most conspicuous, and probably one of the
best documented mechanisms, is that of enhanced grazing pressure from pelagic
zooplankton.
Increased Grazing
Large-sized pelagic zooplankters generally occur in greater numbers inside or
around the edges of macrophyte beds than outside, although this effect can be
modified by fish (see Jeppesen et aI., this volume, Chapter 5, for a comprehensive
discussion). They appear to use the macrophyte beds as a daytime refuge against
fish predation. The concept of this mechanism was first introduced by Timms and
Moss (1984) and has been confirmed or indicated by several other studies (De
Meester et al., 1993; Stansfield et aI., 1995; Lauridsen et aI., this volume, Chapter
13). It is discussed in detail by Jeppesen et al. (this volume, Chapter 5), who also
suggested that there is much higher grazing pressure on phytoplankton inside than
outside the macrophyte beds. An additional consequence of macrophyte presence
for decreasing phytoplankton biomass is that their effects seem to extend beyond
the border of the macrophyte beds. A horizontal diurnal migration of pelagic
zooplankton from dense plant beds covering only 3% of the lake is enough to
increase the grazing potential of zooplankton in the open water by 100% (Lauridsen et aI., 1996).
Grazing by pelagic zooplankton may often be supplemented by grazing from
plant-associated zooplankton species (Irvine et aI., 1989; see also Jeppesen et aI.,
this volume, Chapter 5) and other invertebrate filter and suspension feeders (e.g.,
insect larvae, mollusks). For obvious reasons, plant-associated zooplankters occur
6. Phytoplankton
117
in much higher densities in macrophyte beds than outside. For example, in experiments conducted in Little Mere in 1993, the number of grazers in mesocosms
containing plants was much greater than in those without, with consequent reductions in chlorophyll a concentrations (Beklioglu and Moss, 1996). In addition,
Schriver et aI. (1995), using 100-m2 enclosures in Lake Stigsholm, recorded a
decline in phytoplankton biomass with increasing macrophyte density in terms of
percentage volume infestation (PVI) even when planktonic cladocerans were
absent. They argued that plant-associated zooplankton such as chydorids, Pleuroxus and Eurycercus, which usually were found in tenfold higher densities during
both day and night in the vegetation, could be of significance. Although the effects
on phytoplankton have been rarely documented directly, there are thus reasons to
believe that plant-associated filter feeders are important in decreasing the phytoplankton biomass within macrophyte beds.
Changes in Nutrient Cycling
The presence of macrophytes may in various ways influence nutrient cycling (see
Barko and James, this volume, Chapter 10) and thus potentially also phytoplankton biomass and the growth and competition among different phytoplankters. Decreased availability of nitrogen and sometimes phosphorus may be
expected in most cases.
Macrophytes, and their associated epiphytes, take up and store nutrients, which
are then not available for the phytoplankton (Kufel and Ozimek, 1994; O'Dell et
aI., 1995). A rapid growth of Elodea in Lake Zwemlust was followed by limitation
of nitrogen during summer (Ozimek et aI., 1990), and it was calculated that 64%
of the total lake nitrogen and 61 % of that of phosphorus (excluding the sediment
pool) were found in the macrophyte compartments (van Donk and Gulati, 1995).
Indirect effects are also likely to explain the generally low nitrogen concentrations
often seen in the presence of macrophytes, even when the uptake by macrophytes
themselves is considered low. The structured environment inside macrophyte
beds, which may include gradients leading to anoxic conditions, may increase
denitrification (Weisner et aI., 1994; Jeppesen et aI., submitted a). Changes in
nitrogen competition between phytoplankton and sediment-associated bacteria
may also be important: when phytoplankton biomass is low (in the presence of
macrophytes), the nitrate uptake by the bacteria in the sediment and the subsequent denitrification to N2 is relatively high compared with a situation with a
high phytoplankton biomass (when macrophytes are absent) in which more nitrate
is taken up and stored as organic nitrogen by the phytoplankters. The overall
nitrogen removal is therefore higher in the presence of macrophytes. The general
importance of these mechanisms, however, still needs to be verified, although it is
indicated by several whole-lake studies where the macrophyte density has
changed markedly (van Donk et aI., 1990; Jeppesen et aI., 1997; Moss et al.,
1997).
When phosphorus availability is concerned, the mechanisms are probably more
diverse and the outcome more complicated. Brammer (1979) explained an ex-
118
M. S!/Indergaard and B. Moss
clusion of phytoplankton in the proximity of a dense stand of Stratiotes aloides to
be caused by competition for nutrients together with the co-precipitation of phosphorus with calcium carbonate on the leaf surfaces of the submerged plants. Very
dense macrophyte beds, however, may also increase the availability of phosphorus
(Stephen et aI., 1997). Canopy-forming types, poorly rooted species such as
Elodea and dense Chara beds, which create dense shade, may have more or less
anoxic conditions in their bottom layers (Pokorny et al., 1984; Moss et al., 1986),
thereby increasing the potential for phosphorus release from the sediment, particularly where the iron-bound and redox-sensitive phosphorus fractions constitute
a significant part of the total P in the sediment. This is consistent with observations
in Cockshoot Broad, United Kingdom, in which a recovery of macrophytes was
associated with an increased release of phosphorus from the sediment (Moss et aI.,
1996). An analogous mechanism is an increased phosphorus release from the
sediment due to enhanced pH, caused by high photosynthetic activity, which has
been shown to affect the phosphorus sorption mechanisms in the surface sediment
(S!Ilndergaard, 1988). Plant beds frequently have high pH in the surface layers but
lower values in the darker, deeper layers toward the sediment (Frodge et al., 1990).
The ultimate effects of pH variation in plant beds are thus complex. Finally,
release of organic carbon compounds from macrophytes may also stimulate bacterial production (Wetzel and S!Ilndergaard, this volume, Chapter 7), and thus
indirectly increase bacterial phosphorus uptake at the expense of phytoplankton.
Less dense macrophyte beds and macrophytes having a more developed root
system may, however, affect sediment release rate in a converse way by oxidizing
the surface sediment (Andersen and Olsen, 1994; Flessa, 1994; Christensen and
Andersen, 1996), thereby reducing the redox-sensitive phosphorus release. The
final outcome for phosphorus availability in macrophyte beds may therefore
depend on both macrophyte species and density. In general, however, submerged
macrophytes may be regarded as phosphorus sinks during their active growth and
only as a potential phosphorus source during the relatively short periods of their
senescence (Carpenter and Lodge, 1986; R0rslett et aI., 1986; Jones, 1990).
Although change in nutrient concentrations is probably related to macrophyte
density, it is often difficult to show its effects on phytoplankton biomass because
of masking by other mechanisms acting simultaneously. In their enclosure experiments, Schriver et al. (1995) noted that the concentrations of orthophosphate and
ammonium were strongly and positively correlated with the potential grazing
pressure from zooplankton. They suggested that the shift to low phytoplankton
biomass at high grazing pressure from zooplankton should be attributed to enhanced grazing rather than nutrient control of algal growth. In that particular case,
nutrient changes did not seem to be important. In enclosure experiments in Little
Mere, phosphate and ammonium concentrations were also unaffected by the
presence or absence of macrophytes but strongly influenced by the numbers of
zooplankters, responding to changes in fish predation (Beklioglu and Moss, 1996).
The concentrations of chlorophyll a in the phytoplankton, however, were significantly and positively related to the number of fish both in the presence or
absence of macrophytes.
6. Phytoplankton
119
Allelopathy
Release of organic compounds from macrophytes, in particular from Chara, with
allelopathic effects on phytoplankton has also been suggested to reduce phytoplankton production and biomass (Wium-Andersen, 1982; Gross, 1995; van Donk
and Gulati, 1995; Gross et aI., 1996) or cause changes in the phytoplankton
community structure (J asser, 1995). However, in field experiments using whole
plants and in laboratory experiments using macrophyte extracts, Jasser (1995)
only recorded a minor effect on phytoplankton biomass in the presence of macrophytes or their extracts, although major effects were seen in the community
structure. Similarly, Forsberg et al. (1990) did not find any allelopathic effects of
Chara on chlorophyll a concentrations when comparing phosphorus/Chlorophyll a
relationships in Chara-dominated and non-Chara lakes. Furthermore, they also
noticed a dense epiphytic growth on the Chara plants, which they interpreted as an
absence of allelopathy by Chara in situ. The importance of allelopathic substances, although undoubtedly they exist and can be shown to act in laboratory
situations, remains still to be fully documented at the ecosystem level.
Increased Sedimentation
Increased sedimentation and reduced resuspension, due to the less turbulent and
more quiescent waters among the macrophytes, is yet another factor that may
negatively influence the phytoplankton community (see also Barko and James,
this volume, Chapter 10). In Chesapeake Bay, with semidiumal tides, Kemp et al.
( 1984) recorded a lower phytoplankton biomass in a littoral weedbed area compared with nearby open water. They ascribed this to increased sedimentation
within the weedbed. Similarly, Jones (1990) recorded a consistently lower phytoplankton biomass in the vegetated areas of the freshwater Potomac River than in
the adjacent unvegetated reaches. Van den Berg et al. (this volume, Chapter 25)
recorded a 3D-fold lower phytoplankton density inside dense Chara beds than
outside and attributed it to stiller conditions. In other situations, for example Balls
et ai. (1989), increased sedimentation of phytoplankton in macrophyte beds in
ponds was not found to be important because the phytoplankton community
became dominated by flagellates. The importance of this mechanism may also be
questioned because ofthe very high phytoplankton biomass found in nutrient-rich
brackish lakes, even when macrophyte coverage is substantial (Baks et al., 1993;
Jeppesen et aI., 1994; Jeppesen et aI., this volume, Chapter 28).
Reduced Light Conditions
Finally, there may be shading effects of macrophytes on phytoplankton (Sand-Jensen, 1989; Ozimek et al., 1990). The shading effect is highly related to macrophyte
biomass, but macrophyte surface area also depends on plant species (Sher-Kaul et
aI., 1995), and light extinction coefficients of macrophyte canopies show a fourfold variation among canopy species (review Carpenter and Lodge, 1986). Dense
beds are undoubtedly self-shading (Nielsen and Sand-Jensen, 1989; Frodge et al.,
120
M. S¢ndergaard and B. Moss
1990). However, quite dense periphyton communities may persist within them,
and epipelic algae can be found on the bottom sediment among the plants. The
strategies of buoyancy and motility are available to the phytoplankton community
to overcome this potential problem.
Macrophyte Threshold Effects
In most of the examples mentioned above, the conclusions have been drawn from
comparisons between macrophyte-free and macrophyte-rich areas. An important
aspect, not least from a management point of view, is whether macrophyte density
must attain a certain threshold before they have a significant effect on the phytoplankton. Several studies indicate that this is the case. Canfield et al. (1984) noted
that water transparency was generally high in lakes with a macrophyte PVI
exceeding 30%, and Schriver et al. (1995) found a threshold level of 15-20% PVI
(Potamogeton pectinatus and Potamogeton pusillus), above which a sudden decrease from 30 mm 3JL to 10 mm3JL in phytoplankton biovolume was recorded
(Fig. 6.IA). Later experiments in Lake Stigsholm suggested a relatively low
refuge effect for zooplankton at a PVI below 20-24%, but a significant effect
at a PVI of 50% (Jeppesen et aI., this volume, Chapter 5). The presence of a
threshold and a nonlinear relation with macrophytes is also proposed with
regard to the impact of mollusks (Bronmark and Vermaat, this volume, Chapter
3) and the refuge efficacy for zooplankton (Jeppesen et aI., this volume,
Chapter 5).
Consensus
Although it may be difficult or even futile to attempt to determine which of the
mechanisms mentioned above is decisive, for several may act together, it is
becoming increasingly indisputable that strong antagonistic forces exist between
macrophytes and phytoplankton and that the presence of macrophytes generally is
tantamount to low phytoplankton biomass in freshwater lakes. This statement is
supported by numerous studies comparing phytoplankton in the presence and
absence of macrophytes, including examples from this book (Blindow et al., this
volume, Chapter 26; Faafeng and Mjelde, this volume, Chapter 27; Jeppesen et al.,
this volume, Chapter 28; Van den Berg et aI., this volume, Chapter 25). Balls et ai.
(1989) also concluded that in experimental ponds where the submerged plants
remained intact, plants were able to buffer strongly the effects of added nutrients.
Despite large loadings, nutrient concentrations in the presence of plants could not
build up sufficiently to support large algal growths. By contrast, where the plants
were artificially removed, in a set of replicate but cleared ponds, sufficiently large
concentrations did establish and support large algal crops. In 20-m2 experimental
enclosures in the Danish Lake Stigsholm, a fourfold higher chlorophyll a concentration and a 1O-25-fold higher phytoplankton biomass were recorded in
macrophyte-free enclosures compared with enclosures with a 50% PVI (Table 6.1). The statement is also supported by empirical studies of more than 100
Danish lakes, in which Sec chi disk transparency was, in general, much higher in
121
6. Phytoplankton
A
0
L
"E
E.
0
39
c:
0
!C
c:
26
i5.
13
ra
0
>-
.<::
Cl..
0
i
0
0
-----;,/ ____ ~_O
600
F"-
OJ
~
.!
g
1200 cB
"
0
-;.
Macrophytes (PVI)
B
-Nitzschia
Scenedesmus
Kirschneriella
picoplankton
• C-strategists
• Moderate
sedimentation
susceptibility
• High grazing
susceptibility
.,.
0
-~
l':6.o ...., 0 r--c
o
-
3:
0
0
Ol
3-
0
OU
0
00
0
• Low
sedimentation
susceptibility
0
ra
c:
0
E
en
0
0
0
0
III
ra
-
e
.<::
c.
ra
0
0
o
20
• Moderate to low
grazing
susceptibility
0
00
+ 1200
1800
.Umnothrix
PlanktotIJrix
Microcystis
Aphanothece
Peridinium
Gymnodinium
• S-strategists
D~O
0
0
0
600
~
v 00
40
Macrophytes (PVI)
60
·
·
Cryptomonas
Chlamydomonas
Ankyra
C-strategists
-Low
sedimentation
susceptibility
'High
grazing
susceptibility
Figure 6.1. Phytoplankton biovolume (A) and community structure (B) in relation to
macrophyte PVI (percentage volume infestation) and Daphnia + Bosmina biomass. (Modified from Schriver et al., 1995.)
lakes with a high macrophyte coverage compared with those without plants at
comparable nutrient concentrations (Jeppesen et al., 1990).
Despite a number of reservations and huge differences among lakes and environmental conditions, we have attempted to summarize the effects and make
suggestions concerning the relative importance of different factors associated with
macrophytes on phytoplankton biomass in Figure 6.2.
122
M. S¢ndergaard and B. Moss
Table 6.1. Chlorophyll a Concentrations (Mean of I Daytime and I Night-Time Sampling in 3 Enclosures, ± SD) and Phytoplankton Biomass (Mean of 3 Night-Time and 3
Daytime Samplings in 3 Enclosures ± SD) in Macrophyte-Free (n = 3) and MacrophyteRich (50% PVI, n = 3) 20-m3 Enclosures from July 25 to 27, 1994, in Lake Stigsholm
Without macrophytes
With macrophytes
(n
Chlorophyll a (llglL)
Phytoplankton biomass
(mm3/L)
=3)
(n = 3)
Day
Night
Day
Night
17 (19)
0.80 (0.78)
8.3 (4.0)
0.31 (0.21)
64 (10)
8.1 (1.4)
37 (7.9)
7.6(2.1)
All observations comparing enclosures with and without macrophytes are significantly diferent (P <
.01, t test unequal variance). Only night and day observations of chlorophyll a in macrophyte-free
enclosures are significantly different (P = .02, t test unequal variance).
A. Grazing from
pelagic zooplankton
C/)
C/)
Q)
c
.~
U
Q)
D. Increased sedimentation/
reduced resuspension
~
B. Grazing from
plant associated invertebrates
E. Removed N
~
::::
Q)
OJ
c
'iii
<tI
~
U
C
C. Allelophatic effects
~
F. Removed P
..
Increasing PVI
Figure 6.2. The relative importance of different mechanisms potentially affecting the
phytoplankton biomass at increasing macrophyte PVI (percentage volume infestation).
6. Phytoplankton
123
The importance of grazing by pelagic zooplankton species is probably not
linearly proportional to macrophyte density because of a need for macrophytes to
attain a threshold density before they have a refuge effect (Fig. 6.2A). Furthermore, the macrophyte density required depends on fish density, the refuge effect
being diminished at increasing fish density (Jeppesen et aI., submitted b; Jeppesen
et aI., this volume, Chapter 5). It is also likely that the importance of grazing by
pelagic zooplankton diminishes in very dense macrophyte beds. Daphnia, in
particular, are frequently most abundant at the edges of macrophyte beds rather
than in their interiors (Lauridsen et aI., 1996). As beds cover the lake, edge habitat
is diminished. We think the importance of grazing overall is likely to be high in
most situations, however.
Having exceeded a certain minimum macrophyte density, grazing from plantassociated invertebrates is expected to increase linearly with macrophyte density
(Fig. 6.2B). The effect, however, again may depend on fish predation, although in
mesocosm experiments in Little Mere (Beklioglu and Moss, 1996), there was
no effect of increasing perch (Percafluviatilis) numbers on densities of plantassociated Cladocera (Chydorus, Eurycercus, Simocephalus, Sida) , although
severe reductions of Daphnia numbers, despite the presence of plant effects. At
present, it is difficult to assess the relative importance of plant-associated
grazers as there are few quantitative data and conditions in which such grazing
is likely to be high (in dense plant beds) are also those in which nutrient effects
may be strong.
Allelopathic effects, if present (Fig 6.2C), may be predicted to be proportionately more important at low plant densities, where the plants are growing
vigorously before self-shading and perhaps nitrogen limitation sets in, and to reach
a plateau of effectiveness at moderate macrophyte densities. Bacterial decomposition of the substances is likely to increase as the beds become dense. The effectiveness of allelopathy is also likely to be dependent on macrophyte species. Based on
the present evidence, we suggest that allelopathy is generally of lesser importance
in the ecosystem context than other factors.
Effects of stilling of the water and increased sedimentation and/or reduced
resuspension are expected to increase proportionally with macrophyte density
(Fig. 6.2D), but their overall importance is not well documented. We consider their
importance moderate compared with other factors under most circumstances.
Reduction of nitrogen availability is likely to follow a pattern of increased
effect after attainment of a threshold macrophyte density sufficient to account
for considerable uptake and to create conditions appropriate to denitrification
(Fig. 6.2E). This effect is probably an important one overall, especially in dense
macrophyte beds and in lakes having a relatively low nitrogen loading and consequently the highest possibility of nitrogen limitation of phytoplankton.
The importance of effects on phosphorus will depend more on macrophyte type
and density than most of the other factors. If the macrophyte beds are not too
dense, the effect is likely to be one of reduction in concentration; but if macrophytes become very dense, the effect may be the opposite, with the risk of an
enhanced release from the sediment (Fig. 6.2F). We think that in terms of
124
M. Spndergaard and B. Moss
mechanisms limiting the growth of phytoplankton m weedy, eutrophic lakes,
phosphorus effects are unlikely to be very important.
Fish- and Nutrient-Mediated Effects of Macrophytes on
Phytoplankton Biomass
We have not yet fully considered the influence of fish and nutrients. There are
reasons to believe, however, that fish modify the influence of macrophytes on
phytoplankton. In Chapter 5, it was shown that the behavior of fish and
zooplankton and their interactions were highly dependent on macrophyte density.
We must therefore anticipate consequential effects on phytoplankton. Based on the
threshold level of macrophyte density of 15-20% PVI influencing fish-mediated
impact on zooplankton in the Lake Stigsholm experiments (Schriver et aI., 1995)
and on other results, we have summarized, in Table 6.2, phytoplankton changes
along a nutrient, macrophyte, and fish gradient. Schriver et aI. (1995) argued also
for the existence of a fish threshold. They concluded that the effects on phytoplankton biomass of a high macrophyte coverage were only important as long as
the fish density did not exceed approximately 2 fry/m2. If it was higher, the refuge
effect seemed low and even a high macrophyte density would be unable to have
any major zooplankton-mediated effects on the phytoplankton. These generalizations are probably modified by the actual nutrient level, and it may, for instance,
be possible that nitrogen limitation effects may become more important under
such circumstances because high fish densities can frequently be found in association with dense plant beds.
In Table 6.2, we also suggest how some phytoplankton variables may change
along nutrient, macrophyte, and fish gradients. Such a generalization is complicated by the fact that the macrophyte community itself will also change along a
nutrient gradient. Small vascular plants and Charophyte-dominated communities
are most common at low nutrient concentrations (say, <50 f.lg total PIL). Small,
shallow, and very fertile waters (say, >200 f.lg total PIL) tend to become dominated
by nymphaeids (lilies), with their partly emergent foliage, perhaps because periphyton infestation inhibits maintenance of submerged leaves. Intermediately fertile waters have mixed communities of tall submerged vascular plants, although
charophytes and small vascular plants may still be present and lilies usually
are. Clear water is maintained at all combinations of nutrient concentrations
and macrophyte densities, except low macrophyte density/high nutrients and
high fish densities, by the mechanisms discussed above. Grazing, either by
daphnids or by plant-associated animals, is likely to be important in all circumstances in which plants persist, although the balance of these two grazer
groups will differ. Daphnids may be less important in very dense submerged
macrophyte stands.
The relative importance of the availability of nitrogen and phosphorus is likely
to vary along the gradients (Table 6.2), with phosphorus becoming less limiting
and nitrogen more so as both nutrient loading and macrophyte density increase.
6. Phytoplankton
125
Table 6.2. Generalized Matrix Showing Proposed Effects of Nutrient Concentration,
Fish Density, and Macrophyte Density on Phytoplankton Biomass, Phytoplankton Community Structure, Phytoplankton Cell Size, Transparency, NIP Limitation, and Dominant
Macrophtyesa
Low nutrient
High nutrient
High fish
Low fish
High fish
Low fish
Phytoplankton biomass
Low Macrophytes
High macrophytes
Low
Low
Low/medium
Low
Mediumlhigh
Low/medium
High
Mediumlhigh
Phytoplankton
community
Low macrophytes
Flagellates, zyg
Flagellates, zyg
Diatoms, chlor
Flagellates
High macrophytes
Flagellates, zyg
Flagellates, zyg
Diatoms,
cyano, chlor
Cyano,
flagellates
SmalllIarge
SmalllIarge
SmalllIarge
SmalllIarge
Medium/small
Small
Large/medium
SmalllIarge
High
High
High/medium
High
Mediumllow
Low
High/medium
Low/medium
Phytoplankton cell size
Low macrophytes
High macrophytes
Transparency
Low macrophytes
High macrophytes
NIP limitation
Low macrophytes
High macrophytes
Dominant macrophytes
Low macrophytes
High macrophytes
P
P
NIP
NIP
NIP
NIP
N
N
s. vas/t. vas
s. vasIl. vas
s. vas/t. vas
s. vasIl. vas
1.
t. vaslIiIies
I.
vasllilies
1.
vas/lilies
vas/lilies
Abbreviations: cyano, cyanophytes; chlor, chlorococcales; zyg, zygnemetales; s. vas, small vasculars
and charophytes; 1. vas, tall vasculars.
aApproximate definitions used in this context: low nutrient: P < 25 ~gIL and NIP high; high nutrient:
P> I 00 ~glLand NIP low. Low fish density: <2 fry/m2 ; high fish density: >2 fry/m 2• Low macrophyte
density: PVI < 15%; high macrophyte density: PVI> 30%.
This is because the nitrogen to phosphorus ratios in inflowing waters modified by
human activities, especially those concerning sewage and stock effluents, tend to
be relatively richer in phosphorus than is required for nutrient sufficiency by algae
and because phosphorus is mobilized in dense plant beds, whereas combined
nitrogen is denitrified. Finally, the nature of the phytoplankton is likely to change
along the gradients, with large cells or colonies being favored by increasing
nutrient concentrations, large nitrogen fixers becoming more abundant as nitrogen
stress increases, but small and often flagellate cells and cyanophytes being favored
at high densities of submerged plants due to sedimentation of large nonbuoyant
cells under such conditions.
126
M.
S~ndergaard
and B. Moss
Indirect Effects of Macrophytes on Phytoplankton
Composition and Size
The impact of macrophytes on phytoplankton extends not only to biomass. Macrophytes may also have marked effects on the species and size of individuals
forming the community. As with biomass, several mechanisms may be involved in
determining species composition. Change in the nutrient regime, as mentioned
above, is probably one of the most important. But other factors such as differential
tolerance to shading (Sand-Jensen, 1989), susceptibility to sedimentation in the
darker and quiescent water inside the macrophyte beds, allelopathic effects (Jasser, 1995), and increased input of particulate dissolved organic carbon from the
macrophyte-epiphyte complex, altering the planktonic food chain and thereby
also influencing the phytoplankton, may also be important.
One general tendency is that of increased importance of flagellates in the
presence of macrophytes and/or large-sized zooplankton. For example, Balls et al.
(1989), in fish-free pond experiments, observed no systematic pattern of change
with nutrient loading in either plant-dominated or cleared ponds and a great deal
of variation within the limits of communities all dominated by small and often
flagellated organisms. They attributed this to a stronger selection for small organisms with low sinking rates in the very short water column, compared with any
possible selection for large phytoplankters through grazing. Godmaire and Planas
(1986) in enclosure experiments also recorded a prevalence of flagellates during
the entire season in the presence of Myriophyllum spicatum and suggested that
neither light nor nutrient conditions seemed responsible for this particular community composition. And in Lake Veluwemeer, Van den Berg et al. (this volume,
Chapter 25) recorded a shift in species composition from cyanobacteria to flagellates over a transect from no vegetation to a dense vegetation of Chara.
A tendency toward a greater dominance of small flagellates at increasing
macrophyte abundance, despite an increase in large-sized daphnids, was also
observed by van Donk et al. (1990) in a whole-lake experiment in the biomanipulated Lake Zwernlust (The Netherlands). In this lake, submerged macrophyte
density increased from 0 to 80% within 3 years, and apart from a much lower
phytoplankton biomass, this was accompanied by a shift toward small edible
species. Larger phytoplankton species such as Aphanizomenon, which is considered more resistant to zooplankton grazing, did not develop. This was explained
by dependence of phytoplankton growth on several factors operating simultaneously or successively. Thus, zooplankton grazing in spring, nitrogen limitation
caused by the macrophytes and grazing in summer, and temperature and light
availability in winter controlled phytoplankton growth. It was concluded that
large-sized inedible phytoplankton taxa were unable to oust the small-sized and
edible but fast-growing C strategists, which were more effective in taking up
nutrients during nutrient limitation due to their high surface-to-volume ratio.
A shift to edible cryptophyte flagellates has also been observed in the clear
open water following a fish kill (Samelle, 1993) and in biomanipulated lakes
where the fish stock has been reduced (Leah et al., 1980; Reinertsen et aI., 1990;
6. Phytoplankton
127
S!1Indergaard et aI., 1990). Thus the effects of herbivory on algal succession were
not predictable from the relative susceptibilities of these algal species to grazing
mortality.
In a series of l00-m2 enclosures, Schriver et al. (1995) investigated the importance of zooplankton and macrophyte densities on the phytoplankton composition.
Among their findings was that when Daphnia and Bosmina were abundant,
fast-growing small flagellates such as Cryptomonas and Chlamydomonas dominated. Although cryptophytes (as percentage of biomass) were positively related
to both zooplankton biomass and the interaction between macrophyte PVI and
zooplankton biomass, Chlamydomonas was significantly related only to zooplankton biomass, indicating that Cryptomonas is more dependent on the presence
of macrophytes. In the absence of planktonic cladocerans but in the presence of
macrophytes, cyanophytes (mainly represented by Planktothrix, Limnothrix, Microcystis, and Aphanothece) and dinoflagellates (mainly Gymnodinium and Peridinium) were dominant. Cyanophytes were negatively related to the interaction between
PVI and zooplankton biomass. In the absence of both planktonic cladocerans and
macrophytes, the community was dominated by fast-growing diatoms (mainly
Nitzschia) and chlorophytes (mainly Scenedesmus and Kirchneriella). Chlorococccales were therefore negatively related to PVI*zooplankton biomass. Basically, these findings are in agreement with the life history strategies described by
Reynolds (1987). Fast-growing algae are moderately susceptible to loss by sedimentation and are dominant in the turbulent environment of the macrophyte-free
areas, whereas slow-growing buoyant cyanophytes are dominant in the more
quiescent water among the plants, where the risk of sedimentation is higher.
However, flagellates such as Cryptomonas, with a high grazing susceptibility,
were present at high grazing pressure from zooplankton, and cyanophytes, with
low susceptibility, were present at low grazing pressure, contrary to expectation,
but also often seen in whole-lake studies. On the basis of these findings, Schriver
et al. (1995) argued that the term grazing susceptible should maybe be replaced by
grazing resistant and grazing tolerant, the implication being that grazing-resistant
algae such as Microcystis are only favored until grazing pressure increases to a
certain limit whereafter grazing-tolerant algae such as Cryptomonas take over (see
also Table 6.2). The results of Schriver et al. (1995) have also been summarized in
Figure 6.1B, which defines the three groups of phytoplankton.
It may be asked why small flagellates, vulnerable to zoophmkton grazing, seem
to be such successful competitors in the presence of macrophytes, with the considerable grazer pressure associated with them. A possible explanation (Sommer,
1988) is that flagellates, due to their motility, are better adapted to exploit an
environment that is heterogeneous and structured with respect to nutrient distribution, such as that created by macrophytes. Furthermore, many flagellates are
mixotrophic and able to use DOC excreted from macrophytes and their associated
epiphytic environment. A combination of these advantages and their high growth
rates could be a major reason to their success.
There may also be effects of macrophytes on species composition as well as
changes in the balance of groups such as diatoms, cyanophytes, or flagellates. In
128
M. S!Ilndergaard and B. Moss
600
500
400
300
200
100
a
C':)
E
::L
qj
E
::l
(5
>
c:
ttl
Q)
E
2500
2000
1500
1000
500
Cryptomonas
a
100
Rhodomonas
80
60
40
20
a
a
2
3
4
day
Figure 6.3. Biovolume <11m3) of Chlamydomonas sp., Cryptomonas spp. (c. reflexa, c.
curvata, and C. marssonii), and Rhodomonas sp. in 20-m2 enclosure experiments in Lake
Stigsholm from July 25 (12 AM) to 27 (12 PM), 1994 (sampling at 12 AM and 12 PM), using
macrophyte-free (solid line) and macrophyte-dense (broken line) compartments. Mean
values and standard error of three enclosures with and three enclosures without macrophytes. For total phytoplankton biomass, see Table 7.1.
the Little Mere mesocosm experiments, in which the phytoplankton community
was studied in relation to different densities of fish (Beklioglu and Moss 1996), the
presence or absence of macrophytes had no effect on the total biomass of diatoms
but very significant effects on particular diatom species. Aulacoseira granulata
was much less abundant and Nitzschia palea much more abundant when macrophytes were present.
Results from the Lake Stigsholm experiments, furthermore, indicate that macrophytes can also affect the size distribution at the phytoplankton genus/species
level (Sl1Sndergaard et aI., unpubI.). As an example, marked differences in the size
of the flagellates Chlamydomonas and Cryptomonas were recorded between macrophyte-rich and nonmacrophyte areas, whereas no differences were recorded for
Rhodomonas (Fig. 6.3). Whereas Chlamydomonas sp. was much smaller inside
the macrophyte beds, the mean size of Cryptomonas spp. (three different species)
6. Phytoplankton
129
during the whole experimental period was significantly higher in the presence of
macrophytes (1,100 to 1,800 Ilm3) than in their absence (700-1,000 Ilm3). Differences in grazing susceptibility and size-specific mechanisms to avoid grazing
among different species within macrophyte beds could reinforce the suggestion
that grazing is a major agent affecting the phytoplankton in and around macrophyte beds. The difference in size may also be related to differences in nutrient
concentrations, which are known to influence the size of phytoplankton (Watson
et al., 1992; Hansson and Carpenter, 1993; Larocque et al., 1996). No matter
which mechanism lies behind the results, however, they indicate that macrophytes
can have a structuring effect also on phytoplankton size distribution.
Concluding Remarks
The presence of macrophytes clearly has a negative effect on phytoplankton
biomass. There are indications that along a macrophyte density gradient, the
response is nonlinear, a threshold being recognizable at a macrophyte density of
approximately 15-30% PVI, depending on fish abundance. Several mechanisms
contribute to maintenance of low phytoplankton crops in association with macrophyte beds. These include direct mechanisms associated with the plants themselves, including creation of a still water environment, poor light climate, and
secretion of allelopathic substances as well as mechanisms indirectly linked with
the plants, such as provision of refuges or habitat for grazers on algae, and
modification of the ambient nutrient regime by the metabolic activity of the plants.
In turn, these mechanisms are affected by a second level of influences-predation
by fish on the grazers, imposition of nutrient loads from the catchment, and
hydrodynamic conditions that determine the degree of water movement through
the plant beds.
All these mechanisms interact. The hydrodynamics must determine the extent
to which anoxic gradients that promote denitrification can be established; excretion by grazers modifies the nutrient regime. Furthermore, the influence of specific composition of the macrophyte community is likely to be important. There
are metabolic as well as structural differences among different species of submerged plants, let alone those between, for example, nymphaeids and charophytes. The degree of grazing on algae will also be affected by the composition of
the grazer community, which in turn, is determined by the size structure and
composition of the fish community. The macrophyte growth will also be subject to
grazing by vertebrates so that the structure and composition of the beds will vary.
Major gaps exist concerning the differential effects of different macrophyte
species and of those of mixed fish communities. Most experiments have used
monospecific or simple communities because of the need to balance adequate
replication against the considerable costs of ecosystem experiments, even in
mesocosms. A further gap is that of determining whether allelopathy is important
in the complex ecosystem context. All in all, the mutual interaction of all these
factors makes it highly unlikely that anything other than a very general model of
130
M. S!Ilndergaard and B. Moss
how the macrophyte and phytoplankton communities interact can be constructed. The
importance of different mechanisms and processes will, to some extent, be lakespecific. Much progress has been made in this area through the use of mesocosm
experiments, and this should continue. In addition, accumulating experience from a
variety of different lakes may also begin to reveal greater generalities than we have
been able to discern here. The comparative approach coupled with whole-lake and
mesocosm experiments of increasing scope are likely to be the way forward.
Acknowledgments. The assistance of the technical staff of the National Environmental Research Institute, Silkeborg, is gratefully acknowledged. Phytoplankton
analysis was conducted by B. Laustsen and L. N~rgaard. Technical assistance was
provided by K. M~gelvang, J. Jacobsen, and A.M. Poulsen. The study was supported by the Centre for Freshwater Environmental Research.
References
Andersen, P.0.; Olsen, KJ. Nutrient cycling in shallow, oligotrophic Lake Kvie, Denmark.
Hydrobiologia 275/276:267-276; 1994.
Baks, M.; Moss, B.; Phillips, G.; Irvine, K.; Stansfield, J. The changing ecosystem of a
shallow, brackish lakes, Hickling Broad Norfolk. II. Long-term changes in water chemistry and ecology and their implications for restoration of the lake. Freshwat. BioI.
29:141-165; 1993.
Balls, H.; Moss, B., Irvine, K The loss of submerged plants with eutrophication. 1. Experimental
design, water chemistry, aquatic plant and phytoplankton biomass in experiments carried out in ponds in the Norfolk Broad. Freshwat. BioI. 22:71-87; 1989.
Beklioglu, M.; Moss, B. Mesocosm experiments on the interaction of sediment influence,
fish predation and aquatic plants with the structure of phytoplankton and zooplankton
communities. Freshwat. BioI. 36:315-325; 1996.
Brammer, E.S. Exclusion of phytoplankton in the proximity of dominant water-soldier
(Stratiotes aloides). Freshwat. BioI. 9:233-249; 1979.
Canfield, D.E., Jr.; Shireman, J.w.; Colle, D.E.; Haller, W.T.; Watkins, C.E., II; Maceina,
M.J. Prediction of chlorophyll a concentrations in Florida lakes: importance of aquatic
macrophytes. Can. J. Fish. Aquat. Sci. 41:497-501; 1984.
Carpenter, S.; Lodge, D. Effects of submersed macrophytes on ecosystem processes. Aquat.
Bot. 26:341-370; 1986.
Christensen, KK; Andersen, P.0. Influence of Littorella uniflora on phosphorus retention
in sediment supplied with artificial porewater. Aquat. Bot. 55:183-197; 1996.
De Meester, L.; Maas, S.; Dierckens, K; Dumont, H.J. Habitat selection and patchiness in
Scapholeberis: horizontal distribution and migration of S. mucronata in a small pond. 1.
Plankton Res. 15:1129-1139; 1993.
Flessa, H. Plant-induced changes in the redox potential of the rhizospheres of the submerged vascular macrophytes Myriophyllum verticillatum L. and Ranunculus circinatus
L. Aquat. Bot. 47:119-129; 1994.
Forsberg, C.; Kleiven, S.; Willen, T. Absence of allelopathic effects of Chara on phytoplankton in situ. Aquat. Bot. 38:289-294; 1990.
Frodge, J.D.; Thomas, G.L.; Pauley, G.B. Effects of canopy formation by floating and
submergent aquatic macrophytes on the water quality of two shallow Pacific Northwest
lakes. Aquat. Bot. 38 :231-248; 1990.
Godmaire, H.; Planas, D. Influence of Myriophyllum spicatum L. on the species composition, biomass and primary productivity of phytoplankton. Aquat. Bot. 23:299-308;
1986.
6. Phytoplankton
131
Gross, E.M. Allelopathische Interaktionen zwischen Makrophyten und Epiphyten: Die
Rolle hydrolysierbarer Polyphenole aus Myriophyllum spicatum. Dissertation, Univ. of
Kid Gouingen: Cuvillier Verlag; 1995.
Gross, E.M.; Meyer, H.; Schilling, G. Release and ecological impact of algicidal hydrolyzable
polyphenols in Myriophyllum spicatum. Phytochemistry 41: 133-138; 1996.
Hansson, L.A.; Carpenter, S.R Relative importance of nutrient availability and food chain
for size and community composition in phytoplankton. Oikos 67:257-263; 1993.
Irvine, K; Moss, B.; Balls, H. The loss of submerged plants with eutrophication. II.
Relationships between fish and zooplankton in a set of experimental ponds, and conclusions. Freshwat. Bio!. 22:89-107; 1989.
Jasser, I. The influence of macrophytes on a phytoplankton community in experimental
conditions. Hydrobiologia 306:21-32; 1995.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; S0ndergaard, M.; Mortensen, E.; Sortkjrer, 0.;
OIrik, K Fish manipulation as a lake restoration tool in shallow, eutrophic, temperate
lakes 2: threshold levels, long-term stability and conclusions. Hydrobiologia 200/201:
219-227; 1990.
Jeppesen, E.; S0ndergaard, M.; Kanstrup, E.; Petersen, B.; Eriksen, R.B.; Hammersh0j, M.;
Mortensen, E.; Jensen, J.P.; Have, A. Does the impact of nutlients on the biological
structure and function of brackish and freshwater lakes differ? Hydrobiologia 275/276:
15-30; 1994.
Jeppesen, E.; S0ndergaard, M.; Kronvang, B.; Jensen, J.P.; Svendsen, L.M.; Lauridsen, T.L.
Lake and catchment management in Denmark. In: Harper, D.; Brierley, B.; Ferguson,
A.; Phillips, G.; Madgwick, J., eds. Ecological basis for lake and reservoir management.
London: J. Wiley & Sons (submitted).
Jeppesen, E.; Jensen, J.P.; Sondergaard, M.; Lauridsen, T.; Hald Moller, P.; Sandby, K
Changes in nitrogen retention in shallow eutrophic lakes following a decline in the
density of cyprinids. (submitted a).
Jeppesen, E.; Sondergaard, Ma.; Sondergaard, Mo.; Christoffersen, K.; Jiirgens, K; TheilNielsen, J.; Schltiter, L. Cascading trophic interactions in the littoral zone of a shallow
lake. Limno!. Oceanogr. (submitted b).
Jones, Re. The effect of submersed aquatic vegetation on phytoplankton and water quality
in the tidal freshwater Potomac river. J. Freshwat. Eco!. 5:279-288; 1990.
Kemp, W.M.; Boynton, W.R; Twilley, J.e.; Ward, L.G. Influence of submersed vascular
plants on ecological processes in upper Chesapeake Bay. In: Kennedy, V.S., ed. The
estuary as a filter. Orlando: Academic Press; 1984.
Kufel, L.; Ozimek, T. Can Cham control phosphorus cycling in Lake Lukajno (Poland)?
Hydrobiologia 2751276:277-283; 1994.
Larocque, I.; Mazumder, A.; Prouix, M.; Lean, D.RS.; Pick, F.R. Sedimentation of algae:
relationships with biomass and size distribution. Can. J. Fish. Aquat. Sci. 53: 1133-1142;
1996.
Lauridsen, T.; Pedersen, L.J.; Jeppesen, E.; Sondergaard, M. The importance of macrophyte
bed size for cladoceran composition and horizontal migration in a shallow lake. J.
Plankton Res. 18:2283-2294; 1996.
Leah, R.T.; Moss, B.; Forrest, D. The role of predation in causing major changes in the
limnology of a hypereutrophic lake. Int. Rev. Ges. Hydrobiol. 65:223-247; 1980.
Moss, B.; Balls, H.R.; Irvine K; Stansfield, J. Restoration of two lowland lakes by isolation
from nutrient-rich water sources with and without removal of sediment. J. App!. Eco!.
23:391-414; 1986.
Moss, B.; Stansfield, J.; Irvine, K.; Perrow, M.R; Phillips, G. Progressive restoration of a
shallow lake-a 12-year experiment in isolation, sediment removal and biomanipulation. J. App!. Eco!. 33:71-86; 1996.
Moss, B.; Beklioglu, M.; Carvalho, L.; Kilinc, S.; McGowan, S.; Stephen, D. Vertically
challenged limnology: contrasts between deep and shallow lakes. Hydrobiologia 3421
343:257-267; 1997.
132
M. S0ndergaard and B. Moss
Nielsen, S.L.; Sand-Jensen, K Regulation of photosynthetic rates of submerged rooted
macrophytes. Oecologia 81 :364-368; 1989.
O'Dell, KM.; VanArman, J.; Welch, B.H.; Hill, S.D. Changes in water chemistry in a
macrophyte-dominated lake before and after herbicide treatment. Lake Reserv. Manage.
11:311-316; 1995.
Ozimek, T.; van Donk, E.; Gulati, R.D. Can macrophytes be useful in the biomanipulation
of lakes? The Lake Zwemlust example. Hydrobiologia 200/201 :399-409; 1990.
Pokorny, J.; Kvet; Ondok, J.P.; Toul, Z.; Ostry, I. Production-ecological analysis of a plant
community dominated by Elodea canadensis Michx. Aquat. Bot. 19:263-292; 1984.
Reinertsen, H.; Jensen, A.; Kokksvik, J.I.; Langeland, A.; Olsen, Y. Effects of fish removal
on the limnetic ecosystem of a shallow lake. Can. 1. Fish. Aquat. Sci. 47: 166-173; 1990.
Reynolds, C.S. The response of phytoplankton communities to changing lake environments. Schweiz. Z. Hydrobiol. 49:220-236; 1987.
Rprslett, B.; Berge, D.; Johansen, S.W. Lake enrichment by submersed macrophytes: a Norwegian whole-lake experience with Elodea canadensis. Aquat. Bot. 26:325-340; 1986.
Sand-Jensen, K Environmental variables and their effect on photosynthesis of aquatic plant
communities. Aquat. Bot. 32:5-25; 1989.
Samelle, O. Herbivore effects on phytoplankton succession in a eutrophic lake. Ecol.
Monogr. 63:129-149; 1993.
Schriver, P.; B0gestrand, J.; Jeppesen, E.; S0ndergaard, M. Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large-scale enclosure experiments in a shallow eutropic lake. Freshwat. BioI. 33:255-270; 1995.
Sher-Kaul, S.; Oertli, B.; Castella, E.; Lachavanne, J-B. Relationship between biomass and
surface area of six submerged aquatic plant species. Aquat. Bot. 51: 147-154; 1995.
Sommer, U. Some size relationships in phytoflagellated motility. Hydrobiologia 161: 125131; 1988.
S0ndergaard, M. Seasonal variations in the loosely sorbed phosphorus fraction of the
sediment of a shallow and hypereutrophic lake. Environ. Geol. Wat. Sci. 11: 115-12 I;
1988.
S0ndergaard, M.; Jeppesen, E.; Mortensen, E.; DaII, E.; Kristensen, P.; Sortkjler, O. Phytoplankton biomass reduction after planktivorous fish reduction in a shallow, eutrophic lake: a
combined effect of reduced internal P-Ioading and increased zooplankton grazing. Hydrobiologia 200/20 1:229-240; 1990.
Stansfield, J.H.; Perrow, M.R.; Tench, L.D.; Jowitt, A.J.D.; Taylor, A.A.L. Do macrophytes
act as refuges for grazing cladocera against fish predation? Wat. Sci. Techn. 32:217220; 1995.
Stephen, D.; Moss, B.; Phillips, G.L. Do rooted macrophytes increase sediment phosphorus
release? Hydrobiologia 342/343:27-34; 1997.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
van Donk, E.; Gulati, R.D. Transition of a lake to turbid state six years after biomanipulation: mechanisms and pathways. Wat. Sci. Techn. 32: 197-206; 1995.
van Donk, E.; Grimm, M.P.; Gulati, R.D.; Klein Breteler, I.P.G. Whole-lake food-web
manipulation as a means to study community interactions in a small ecosystem. Hydrobiologia 200/20 1:275-289; 1990.
Watson, S.; McCauley, E.; Downing, I.A. Sigmoid relationships between phosphorus, algal
biomass, and algal community structure. Can. J. Fish. Aquat. Sci. 49:2605-2610; 1992.
Weisner, S.; Eriksson, G.; Graneli, W.; Leonardson, L. Influence of macrophytes on nitrate
removal in wetlands. Ambio 23:363-366; 1994.
Wium-Andersen, S.; Christophersen, c.; Houen, G. Allelopathic effects on phytoplankton
by substances isolated from aquatic macrophytes (Charaies). Oikos 39:187-190; 1982.
7. Role of Submerged Macrophytes for
the Microbial Community and Dynamics of
Dissolved Organic Carbon in Aquatic Ecosystems
Robert O. Wetzel and Morten S0ndergaard
Introduction
Examination of the multi faced functions of submerged macrophytes in shallow
lakes has once again placed emphasis on the habitat characteristics of the vegetation for fish and motile invertebrate communities. The habitat support functions of
the macrophytic community are critical to these animals for refuge and related
cryptic behavioral functions. We argue here, however, that much more fundamental structuring of microbial metabolism and biogeochemical cycling of the ecosystem result from the development of submerged macrophytic communities.
These metabolic functions not only control biogeochemical cycling within these
lake ecosystems but are essential to the success of diverse integrated higher
trophic levels.
Plant Growth and Habitat Characteristics
The general productivity gradient across the land-water interface clearly recognizes that the maximum productivity of the biosphere occurs in the zone of
emergent macrophytes (e.g., Westlake, 1963; Wetzel, 1990). These plants have
several morphological, physiological, growth, and reproductive adaptations that
capitalize on the high availability of water and nutrients and the suppression of
competitive interactions with many less tolerant plant species.
133
134
R.G. Wetzel and M.
S~ndergaard
Macrophytic productivity declines precipitously when submerged. Ample
physiological evidence indicates that the primary suppressing mechanisms are the
exponential light attenuation with depth and the reduction of gas and nutrient
exchange, particularly for inorganic carbon, largely because diffusion processes
for gases are some 104 times slower than in air (e.g., Raven, 1984). Although,
theoretically, the exchange processes. should be influenced greatly by boundary
layer characteristics, the macrophyte beds themselves change the flow rates and
the flow characteristics, both as a function of density and species. In addition, the
presence of attached epiphytic microbial communities complicate and increase the
diffusive layers and both physically and metabolically separate the macrophyte
surfaces from surrounding water (Losee and Wetzel, 1993; Wetzel, 1993a, 1996).
Thus, two common conditions emerge. (1) External water flow rates are rapidly
dissipated within a few centimeters of the outer plant-bed boundary, even under
severe external flow-rate conditions (30 cmls or more), particularly among submerged macrophytes with large surface areas per unit biovolume (Losee and
Wetzel, 1993). As a result, within-bed flows are so low that boundary layers are
many millimeters in thickness and totally diffusion dependent. The same reduction
is also found in macrophyte beds within moderately flowing streams (Madsen and
Warncke, 1983; Sand-Jensen and Mebus, 1996). (2) Chemical conditions within
periphyton communities are not only considerably different from the surrounding
waters but highly dynamic on short time scales (minutes) (e.g., Carlton and
Wetzel, 1987, 1988; Riber and Wetzel, 1987). These epiphytic communities are
frequently viewed as resource competitors for the submerged macrophytes (e.g.,
Phillips et al., 1978), and indeed they can be when excessively developed under
very eutrophic conditions. However, from the standpoint of the ecosystem, the
productivity is effectively shifted from the macrophytes to the epiphytic communities on the submerged macrophytes. The macrophytes, although not profoundly productive, survive and are essential in providing substrata and massively
diverse three-dimensional habitats for microbiotal colonization. As discussed from
a mechanistic viewpoint in greater detail below, the result is a shift in productivity
from the macrophytes to the very high productivity of the attached microbiota.
Certain morphological and physiological adaptations occur in submerged angiosperms, such as marked increases in surface area and pigment concentrations,
as well as particularly efficient recycling of gases and nutrients within gas lacunae
distributed throughout foliage and rooting tissues (cf. Sondergaard and Wetzel,
1980; Wetzel et aI., 1984). Two important growth characteristics emerge in regard
to submerged macrophytes that are directly germane to their roles in shallow lake
ecosystems. (1) First, from the standpoint of the ecosystem, the most successful
and productive submerged macrophytes are the herbaceous perennials that
develop highly dissected foliage within the water column and not the rapidly
growing surface canopy-forming species such as Myriophyllum and Hydrilla
under conditions of eutrophy. Productivity of rosette perennials such as the isoetids are very much less productive, as are most of the few annual submerged
plants. The most abundant emergent, floating-leaved, and submerged aquatic
macrophytes are herbaceous perennial plants. True annual submerged plants,
7. Dynamics of Dissolved Organic Carbon
135
germinating from seed and essentially returning to seed biomass at the end of a
brief growing season, constitute less than 10% of aquatic plants (Hutchinson, 1975).
Two important ecosystem characteristics emerge from this characteristic. (a) Much of
the nutrient pool is significantly recycled by translocation to the rooting tissues as
foliage of each cohort completes growth. Nutrient recycling is further enhanced by
acquisition of nutrients by the plants largely from interstitial waters of the hydrosoils, and most of it from microbial degradation of residual plant tissue and
epiphytic microorganisms. (b) These herbaceous perennials exhibit continuous
growth with numerous overlapping cohorts. For example, the emergent rush
]uncus effusus has five cohorts per year (latitude 33°; Wetzel and Howe, in press)
and Typha latifolia three cohorts per year (latitude 42°; Dickerman and Wetzel,
1985), and the rooted floating-leaved Nymphaea odorata has seven per year
(latitude 33°; Carter, 1995). The submerged plant Scirpus subterminalis has three
cohorts per year (latitude 42°; Rich et aI., 1971) and Ceratophyllum demersum
four to six cohort turnovers per year (latitude 42°; Spencer and Wetzel, 1993).
Both of the latter species overwinter under ice cover (ca. 4 months) in viable
evergreen, basal metabolic condition, which gives them great advantage in acquisition of nutrients during winter periods when nutrients are relatively abundant
and rapid growth in the spring before that of competitors. Therefore, the variable
but continuous growth results in continuous turnover of organic matter with
production of relatively resistant particulate organic matter in senescent tissues
and provides massive and constantly regenerating surface areas for photosynthetic
and heterotrophic microbial colonization over the entire year. Annual plants do not
exhibit such continuous growth and biomass turnover, and some perennials do not
overwinter in large accumulations of aboveground biomass. Where examined in
detail (e.g., references cited above; also Wetzel, 1983a), however, many of the
submerged macrophytes exhibit high rates of turnover and, importantly, regeneration of surfaces for epiphytic microflora.
(2) Second, the thin, finely divided, and reticulated foliage increases surface
area, which can greatly enhance exchange of gases with those of the water and
interception of light (Sculthorpe, 1967). The resulting high ratio of surface area to
volume increases substrata available for colonization by epiphytic algae, cyanobacteria, and other microbes. For example, leaf surface area available for colonization by epiphytic microbes on the submerged linear-leaved macrophyte Scirpus
subterminalis averaged 24 mZ/mz of lake area in the moderately developed littoral
zone of a lake in southwestern Michigan (Burkholder and Wetzel, 1989). Such
values are likely conservative and in reality can be much larger with even more
finely dissected macrophytes. The extensive area of these submerged macrophyte
surfaces provide myriad diverse microhabitats with attendant attached algal!
cyanobacterial communities upward into littoral environments of relatively abundant light and dissolved gases from photosynthesis (Oz) and decomposition (CO z).
Nutrients within senescing macrophytes and their epiphytic microflora are displaced to the sediment surfaces and recycled. Some nutrients diffuse from the sites
of decomposition to the interstitial waters of the sediments and detrital particulate
matter of macrophytic origins. As a result, the productivity of epiphytic algae/
136
R.G. Wetzel and M. S0ndergaard
cyanobacteria and of benthic algae on macrophytic detritus frequently exceeds
that of the supporting macrophytes (Wetzel, 1983b, 1996). Very high algal and
microbial diversity (e.g., >80% of the freshwater species of algae; Round, 1981)
are associated with highly dynamic and productive microhabitats of the surfaces
of submerged macrophytes.
Loading and Fates of Dissolved Organic Carbon
Bacterioplankton
A widely held opinion in limnology is a direct quantitative coupling of bacterial
productivity to phytoplankton productivity (Cole et aI., 1988). Loadings of dissolved organic carbon (DOC) to shallow aquatic ecosystems are manifold, however, from both external and numerous internal sources of photosynthesis and
decomposition (discussed further below). Most of the loadings of organic carbon
from terrestrial and wetland sources is as DOC, which although certainly not
refractory contains many compounds that are recalcitrant to rapid degradation
(e.g., Wetzel, 1995). In addition, many other sources of DOC enter the pelagic
zone from littoral productivity and from decomposition of imported organic matter. Thus, the dependence of bacterioplankton on DOC released from phytoplankton is highly variable. In every case in which the sources of DOC for
bacterioplankton have been examined carefully in lakes, DOC produced by phytoplankton is inadequate to support bacterioplanktonic metabolism and growth by a
factor of 3 to as much as 20 in shallow lakes (e.g., Borsheim and Andersen, 1987;
Hessen, 1992; Tranvik, 1992; Spndergaard, 1993; del Giordio and Peters, 1994;
Coveney and Wetzel, 1995). The rates of nonphytoplanktonic particulate organic
carbon being mineralized by bacterioplankton can be modified strongly by grazing
interactions of protozoan and metazoan grazing (e.g., Lyche et al., 1996).
Submerged macrophytes function in two major ways in regard to the impact of
microheterotrophs and DOC turnover in the water column: (1) as a source of DOC
for bacterioplanktonic productivity (discussetl below), and (2) as a refuge for
mesozooplankton from vertebrate predation, which allows much greater population densities of cladocerans and their grazing on flagellates, ciliates, and other
protists. Although the planktonic protists have little direct link to submerged
macrophytes, Daphnia can be very effective in reducing nanoflagellates and
ciliates (e.g., to as much as 70% of planktonic nanoflagellate biovolume; Christoffersen, this volume, Chapter 17). As has been shown often, Daphnia populations
can be very large as they develop in the food-abundant predation refuge of
submerged macrophytes (Jeppesen et al., this volume, Chapter 5). Many of the
Daphnia migrate laterally from dense to less densely colonized areas of the littoral
or into the pelagic zone at night to feed on phytoplankton and protistian communities in the predation-reduced time window. Bacterioplankton abundance is
low, but specific productivity can be high under these periods of heavy mesozooplankton grazing (S0ndergaard et aI., this volume, Chapter 15), but the bacteria
7. Dynamics of Dissolved Organic Carbon
137
can compensate with morphotype changes induced by high grazing pressures
(Jurgens and Jeppesen, this volume, Chapter 16). Grazing-resistant morphotype
changes are found under high nanoflagellate and ciliate grazing and not when the
bacterioplanktonic grazing is dominated by mesozooplankton. When submerged
macrophytes are removed or suppressed, fish predation reduce Daphnia populations to low levels (e.g., <1OIL) and a corresponding increase in nanoflagellates
and other protists. The latter microheterotrophs cycle carbon among the bacteria,
and relatively little is incorporated into zooplankton and higher trophic levels
(Lyche et al., 1996).
Bacterioplanktonic couplings with protistian grazing and their interactions with
higher trophic levels can sometimes be effective in regulating bacterial productivity in the pelagic zones during some productive periods of the year. Although
the macrophytes are functioning importantly as DOC sources for bacterioplankton
and as refuge areas for influencing grazers on phyto- and bacterioplankton, we
contend that with the development of submerged macrophyte communities, bacterial productivity shifts from moderate levels within planktonic communities to a
dominance by attached bacterial productivity. The greatly increased attached
bacterial productivity results not only from the large habitat development provided
by the macrophyte surfaces but because of the metabolic couplings between
attached algae/cyanobacteria and the attached bacteria. As a result, the primary
mechanisms of nutrient retention and recycling are shifted from planktonic to
sessile microbiota.
Attached Bacteria of Surfaces
In every lake examined in detail with quantitative measures of rate functions (not
simply correlations), and particularly in shallow lakes, DOC obtained from phytoplanktonic sources (extracellular release, lysis) is totally inadequate to support the
observed rates of even bacterioplankton productivity (e.g., Hessen, 1992; Tranvik,
1992; Coveney and Wetzel, 1995; SS'lndergaard et aI., this volume, Chapter 15) or
are not related directly to phytoplankton (Jeppesen et aI., 1992; SS'lndergaard,
1993). The submerged macrophytes have heavily colonized surfaces, and the
summation of the bacterial productivity on a three-dimensional basis is very much
greater than are those rates in the bacterioplankton. We hypothesize here that most
of the bacterial productivity is associated with suifaces, particularly those of
submerged macrophytes and particulate organic detritus of these plants as they
are constantly growing and cohorts are turning over with the detrital mass collecting at
the sediment-water interface. Although the conceptual framework of the biofilmmacrophyte-pelagic interactions was proposed many years ago (Wetzel and Allen,
1970) and variously analyzed and verified (e.g., Allen, 1971; Allanson, 1973),
good quantitative measurements of attached bacterial productivity are very few,
difficult because of methodological problems caused by the large diffusive boundary layer, and inadequate for extrapolation (Table 7.1). A few other examples exist.
In Lake Stigsholm, attached bacterial productivity averages ca. 1,000 mg C/m2/
day on leaves of Potamogeton pectinatus at 24 m2/m2, whereas the bacterioplanktonic
138
R.G. Wetzel and M. S!2lndergaard
Table 7.1. Bacterial Abundance and Production on Submerged Macrophytes
Species
Zostera marina"
Abundance
(celUcm2)
Production
(J..lg C cm-2 h- 1)
8 x 107
0.14
Z. marina
Green leaves
Brown leaves
Z. marina
2x 107
4-6 x 107
107
2.8
4-6
Z. marina
2-8 x 106
0.4
z. capricomib
5.2 X 107
1.5 X 107
0.005
0.2
Cymodocea nodosa"
Laminaria pallida
103_107
Ecklonia maximata
106
103_107
Laminaria
longicruris
Macrocystis
integrifolia
Rhizophora mangle
Ranunculus
penicillatus
2-37 x 106
4.6 X 106 cells
1.6
2.6 x 107
funcus effusus
FDC
References
Blum and Mills
(1991)
Newell (1981)
Kirchman et al.
104_105
2.5 x 106
2.2 x 108
105
Remarks
TIl, Initial
decomposition
0.06
TIl, average
TIl
From 02
respiration
Min-max season
average for tip
(1980)
Kirchman et al.
(1984)
Moriarty et al. (1985)
Peduzzi & Herndl
(1994)
Mazure & Field
(1980)
Highest viable
counts
Seasonal plate
counts
Young to
senescing
blades
TTl, day 1 day 6
Mazure & Field
(1980)
Laycock (l974)
Leaf I, direct
count
Leaf 6,
tropological
variation
LEU, average
for plant
detritus
Hossell & Baker
(1979)
Velji & Albright
(1986)
Benner et al. (1988)
Thomaz & Wetzel
(1995)
(4-32/gDW)
"z. marina: 350 cm2/g DW, same for Cymodocea.
value applies to sediment area; leaf are index is unknown.
Abbreviations: FDC, frequency of dividing cells; TIl, 3H-thymidine incorporation; LEU, 3H-Ieucine
incorporation.
b The
productivity was ca. 300 mg C/m2/day (Mo. SjZlndergaard, unpublished data). In
another small Danish lake, Dystrup, heavily colonized by Ceratophyllum submersum, estimates in July and August of attached bacterial productivity (leucine to
protein methodology) was ca. 68 g C/m2/48 days versus ca. 21 g C/m2 by the
bacterioplankton over the same time period (Theil-Nielsen and SjZlndergaard,
1997). Even though the actual rates per unit area are often modest, at least with
the conservative methods used, the total area in three-dimensional water column
7. Dynamics of Dissolved Organic Carbon
139
RELATIVE BACTERIAL
PRODUCTION / AREA
BACTERIOPLANKTON
EPIPHYTIC ON SUBMERSED
MACROPHYTES
EPI· AND WITHIN
SEDIMENTS
Figure 7.1. Suggested relationships of bacterial production per unit of lake area at surface
in three habitats of shallow lakes with submersed macrophytes. (A) Bacterioplankton =
production per m2 of water column per year; (B) epiphytic on submersed macrophytes =
production on all plant surfaces within the water column greater than I m2 of sediment per
year; (C) epipelic and within sediments =production per m 2 of sediments per year.
is very large. Hence, the collective bacterial productivity per square meter of water
column below each square meter of water surface can be very large. A depth-scaling factor for the bacterioplankton is of equal importance, as is the leaf-area
scaling factor within the water column for the macrophytes.
Because the productivity of the emergent, floating-leaved, and submerged
macrophytes is high and constantly turning over, leachate from both living and
senescent aquatic macrophytes is high and supports large communities of microbiota. Much of this leachate is used efficiently (~5%/day) by the attached
microbial community; relatively recalcitrant DOC compounds are released downgradient and used at slower rates (ca. 1%/day) (Otsuki and Wetzel, 1974; Mickle
and Wetzel, 1978, 1979; S0ndergaard, 1983, 1990; Moran and Hodson, 1989;
Wetzel, 1992; Mann and Wetzel, 1995, 1996). Partial photolytic degradation of
recalcitrant dissolved organic compounds from allochthonous and from macrophytic sources occurs by natural ultraviolet (UV-B) of sunlight (cf. Wetzel et aI.,
1995; Moran and Zepp, in press). These photolytic products from humic and fulvic
substances include many low-molecular-weight carbonyl compounds, such as
numerous fatty acids, and nutrients including NH! and Po.l-, which are readily
used and enhance bacterial productivity.
140
RG. Wetzel and M.
S~ndergaard
Thus, most bacterial metabolism in shallow lakes with submerged macrophytes
is associated with benthic surfaces, as was emphasized long ago (Wetzel et al.,
1972; Wetzel, 1983a). We project here that a major attribute of shallow lakes in
which submerged macrophytic communities are well developed is the major
increase in bacterial productivity associated with macrophytic surfaces (Fig. 7.1).
The submerged macrophytes function as a major source of organic carbon, but a
primary function is provision of three-dimensional surfaces within the water
column. The bacteria are directly coupled to the productivity of the epiphytic
algae. One cannot view P:R ratios of the phyto- and bacterioplankton alone; this
myopic evaluation must be expanded to include the collective bacterial metabolism within the water column and that associated with surfaces (i.e., an ecosystem
evaluation). When this corrected perspective is done, the P:R ratio is always less
than 1 because of the importation of large amounts of particulate and especially
DOC from allochthonous and land-water interface sources. Most of that respiration is detrital-based and associated with surfaces, particularly benthic surfaces
(Wetzel, 1995).
Pivotal Metabolic Roles of Epiphytic Microbiota
For many years, microbiota attached to surfaces of living and detrital organic
materials in littoral areas and wetlands have been largely ignored as quantitatively
insignificant. Supporting evidence is still sparse but overwhelming that these
microbial communities of autotrophic algae and cyanobacteria, and bacteria,
fungi, and other heterotrophic microorganisms are not only very productive but
can often serve as major regulators of nutrient dynamics in many freshwaters
(Wetzel et aI., 1972; Wetzel, 1983a,b, 1990). The importance of epiphytic algal
productivity of Lawrence Lake, southwestern Michigan, is taken as an example
because it has been studied in detail for many years. Submerged macrophytes
extend in dense beds from 0 to 5.3 m of depth but constitute less than one-quarter
of the annual net primary productivity in the lake of 180 g C/m 2 (Table 7.2). The
rest originates from phytoplanktonic (13%) and epiphytic (71 %) algae (Burkholder and Wetzel, 1989). These ratios are likely very common among shallow
oligo- to moderately eutrophic lakes and certainly are not exceptional. Very few
lakes have been studied in the required detail to make such comparisons (S0ndergaard and Sand-Jensen, 1979).
The exceptionally high rates of primary productivity of the attached algae and
cyanobacteria are only possible because of the intensive internal recycling of
nutrients, including carbon, and gases within the attached microcommunities
(Wetzel, 1993a). Steep gradients and very large boundary layers (millimeters in
thickness) exist between the overlying water and within these attached microbial
communities embedded in dense mucopolysaccharide matrices. The encapSUlation of polysaccharide matrices increases densities and reduces diffusion rates
even further. To maintain the very high ratio of attached algal productivity always
measured in these communities, rapid, intensive recycling of inorganic nutrients
141
7. Dynamics of Dissolved Organic Carbon
Table 7.2. Reconstructed Carbon Budget for Primary Production in Lawrence Lake,
Considering Epiphytic Algae from Depths 0--5 m"
Component
Phytoplankton
Littoral zone algae <1 m
Littoral zone algae 1-5 m
Macrophytes ::;5 cm
Total
Mean daily
(mg C/m2/day)
Mean annual
(kg C/lake/yr)
Contribution
119
2,001
500
241
2,154
5,512
5,968
2,701
16,335
13
34
37
16
100
(%)
"Epipelic algal productivity was evaluated and constitutes <I % of the total in this lake (Carlton and
Wetzel, unpublished).
(Adapted from data of Wetzel et al. [1972]. Rich et al. [1971], and Burkholder and Wetzel [1989].)
and particulate and dissolved organic matter must be occurring among the producers and the heterotrophic organisms (bacteria, fungi, and protists) within the
periphyton. This mutualistic recycling, demonstrated experimentally (Neely and
Wetzel, 1995), allows the communities to maintain themselves and to efficiently
sequester external sources of nutrients from the overlying water or from the
substrata on which they grow for net growth, export, and reproduction.
Critical to the efficiency of the internal recycling is the coupling between
attached microbial photosynthesis and heterotrophic bacterial metabolism. Rapid
recycling of nutrients and gases within the periphytic complex is facilitated by the
proximate juxtaposition of cells, often in immediate contact with each other.
Experimental analyses have demonstrated a tight metabolic coupling among these
organisms. For example, suppression of photosynthesis of the attached algae
results in an immediate reduction of attached bacterial productivity (Neely and
Wetzel, 1995) and certainly also nutrient and gas recycling. As a result, the
nutrient retention capabilities of the attached communities is reduced markedly if
the photosynthetic interactions within these mutually coupled communities are
reduced or removed.
Extracellular mucilaginous material (i.e., exopolysaccharides or exopolymer
secretions) occurs as coatings around most individual microbial cells or projections from cells. This material ultimately forms a matrix inhabited by a variety of
microorganisms, particularly bacteria, algae, protozoa, and fungi (e.g., Fletcher
and Marshall, 1982). Excretion of exopolymer fibrils by bacteria is an important
initial phase of attachment of microbes to surfaces (Fletcher and Floodgate, 1973;
van Loosdrecht et aI., 1990; Fletcher, 1991), and a substantial portion of any
attached microbial community will be composed of nonliving mucilaginous
materials. Most of this material is mucopolysaccharide, although the exact composition and texture varies with environmental conditions as well as the nature and
condition of the organisms that secrete it (e.g., Sutherland, 1985).
Problems of slow nutrient diffusion into attached microbial communities are
particularly acute in relatively thick communities, where high productivity is
142
R.G. Wetzel and M.
S~ndergaard
maintained by intensive internal recycling of nutrients and gases among the
metabolically interdependent microbiota (Wetzel, 1993a). Oxygen and CO 2 are
interchanged from photosynthesis and respiration, DOC, and organic micronutrients (e.g., vitamins) released from algae or bacteria-large portions are exchanged rapidly without leaving the confines of the attached communities.
Certainly, the metabolism of the supporting macrophyte functions both as a source
and recipient of materials of the attached community. For example, some epiphytic algal species growing adnate to submerged macrophyte leaves can obtain
more than 60% of their phosphorus from the living macrophytes (Moeller et aI.,
1988). Similarly, adnate bacteria can obtain significant carbon from the macrophyte (Kirchman et aI., 1984).
Abiotic adsorption of particulate and dissolved organic carbon, as well as
incorporation of inorganic nutrients and metals into the mucilaginous matrix, can
provide a supplemental mechanism to diffusion that results in nutrients/metal
inputs into attached microbiota (Lock, 1981, 1990; Wetzel, 1983b; Lock et aI.,
1984; Roemer et aI., 1984; Beveridge and Graham, 1991). This mechanism likely
not only stimulates metabolism of the community but, in the case of certain cations
(e.g., Ca, Mg, Fe, Mn), affects the physical properties of the mucilage matrix, such
as hydrophobicity (e.g., Lemke et aI., 1995) and possibly texture. The extracellular
matrix also provides sites of attachment for extracellular enzymes (e.g., phosphatases and proteases) that are critical in rendering nutrients and carbon available
to microorganisms (Wetzel, 1990, 1991).
Further studies have examined the effects of varying concentrations of natural
dissolved organic acids from decomposing aquatic plants on the metabolism and
productivity of attached algae and bacteria (Wetzel, 1991, 1992; Kim and Wetzel,
1993; Wetzel et aI., 1997). Microscopic examination of periphyton communities
during these studies revealed apparent reductions in and alterations of mucilaginous matrices on exposure to various types of natural and purified organic acids
from decomposing macrophytes. Direct experiments, using both scanning electron
microscopy and energy dispersive X-ray microanalyses of copper-containing dyes
that bind to glycosaminoglycans of the mucilage, showed that DOC-treated
communities contained 10-57% less mucilage than non-DOC-treated controls
(Wetzel et aI., 1997). These examples only suggest (1) the importance of the
mucopolysaccharide matrices in influencing the rates of adsorption and transport
of nutrients and gases into and from the periphytic communities and (2) that many
environmental parameters potentially influence the diffusion rates into the communities from the overlying water. These parameters are likely dynamic in their
effects on mucopolysaccharide permeability properties on a short time basis.
Conclusions
1. In addition to habitat functions critical for refuge and behavior of fish and
invertebrates, we argue that submerged macrophyte-epiphytic microbiotic communities function in fundamental structuring of microbial metabolism and
7. Dynamics of Dissolved Organic Carbon
2.
3.
4.
5.
143
biogeochemical cycling of entire shallow lake ecosystems. Additional structuring roles encourage development of mesoplankton and depression of phytoplankton and microbial protista. Bacterial productivity is increased by high
rates of nutrient recycling and the metabolic couplings within the macrophyteperiphyton complex.
The instantaneous primary productivity of submerged macrophytes is moderated
because of a number of physical and physiological constraints, particularly reduced light and gaseous/nutrient diffusion in water. These environmental limitations have resulted in (1) a predominance of continuously growing and senescing
herbaceous perennial species with a number of multiple, simultaneously growing
cohorts at different stages of development and (2) a marked increase in surface
areas of living, senescing, and dead tissues.
The extremely high surface area of living, senescing, and dead tissues projecting within and up into the water column promotes development of a highly
mutualistic attached microbial community. Preliminary experimental data suggest that these attached communities require rapid intensive recycling of carbon, phosphorus, nitrogen, and other nutrients and growth factors between
producers, particulate and dissolved detritus, and bacteria and protists. The
organization of the attached communities in a matrix of mucopolysaccharides
in a diffusion controlled environment facilitates internal community recycling
and results in a highly efficient reutilization of resources. This high growth can
then augment the internal recycling and small losses with external inputs of
carbon and nutrients from the surrounding water or from the supporting substrata. As a result, a shift occurs in the ecosystem productivity from submerged
macrophytes of modest productivity to very high productivity of the attached
microbiota. Attached algal productivity can exceed that of the submerged
macrophytes and phytoplankton combined. Composite attached bacterial productivity also can be much greater than that of the bacterioplankton and can
dominate ecosystem degradative productivity.
DOC from allochthonous sources entering the submerged macrophyteepiphyte complex can be effectively scavenged by physical and metabolic use.
Use of recalcitrant DOC can be markedly enhanced by partial UV-B photolysis
by natural sunlight. Rapid use of much of the DOC from this photolytic
decomposition and that of senescing submerged macrophytes results in selective increases in the chemical recalcitrance of DOC passing through the submerged macrophyte-detritus-epiphyte complex. The P:R ratios for the lake
ecosystems are always less than 1 because of the importation of large amounts
ofDOC.
Productivity and nutrient recycling capacities are maximized within the
water by extensive development of submerged macrophytes and associated
epiphytic microbiota. Prevention of the development of the submerged
macrophyte-epiphytic complexes by any mechanism (e.g., turbidity, excessive nutrient enrichment, epidemic herbivory, toxicity) will usually result in
precipitous declines in productivity, nutrient recycling, and nutrient retention capacities.
144
R.G. Wetzel and M. S!Ilndergaard
We recognize that these conclusions are drawn from relatively few data. Much
additional supporting information is needed, and we particularly recommend
intensive investigation of the following:
• Intensive comparative quantitative measurements of microbial production in
attached communities against those of the open water column. These measures
are complicated by the slow diffusion rates and rapid recycling that occur
within the attached communities, as discussed in this chapter.
• The sources of DOC used by attached and planktonic bacteria from allochthonous, macrophytic, and phytoplanktonic sources. It is particularly
important to evaluate the suggestions made in this discourse that most
submerged macrophytes do not senesce en mass but rather are constantly
turning over with multiple continuous cohorts .
• The relative impacts on bacterial productivity of DOC from submerged
macrophytes released rapidly from cells undergoing slow, progressive senescence versus DOC released from attached bacteria and fungi that are slowly
degrading the structural tissues of submerged macrophytes.
• If the presence of submerged macrophytes in general promote the development of large populations of mesozooplankton and suppress protistian micrograzers and thereby reduce the effectiveness of microbial loop cycling in the
water.
References
Allanson, B.R. The fine structure of the periphyton of Chara sp. and Potamogeton natans
from Wytham Pond, Oxford, and its significance to the macrophyte-periphyton metabolic model ofR.G. Wetzel and H.L. Allen. Freshwat. BioI. 3:535-542; 1973.
Allen, H.L. Primary productivity, chemo-organotrophy, and nutritional interactions of epiphytic algae and bacteria on macrophytes in the littoral of a lake. Ecol. Monogr.
41:97-127; 1971.
Benner, R.; Hodson, R.E.; Kirchman, D. Bacterial abundance and production on mangrove
leaves during initial stages of leaching and biodegradation. Arch. Hydrobiol. Beih.
Ergebn. Limnol. 31: 19-26; 1988.
Beveridge, T.J.; Graham, L.L. Surface layers of bacteria. Microbiol. Rev. 55:684-705;
1991.
Blum, L.K.; Mills, A.L. Microbial growth and activity during the initial stages of seagrass
decomposition. Mar. Ecol. Progr. Ser. 70:73-82; 1991.
Borsheim, K.y.; Andersen, S. Grazing and food selection by crustacean zooplankton
compared to production of bacteria and phytoplankton in a shallow Norwegian mountain lake. J. Plankton Res. 9:367-379; 1987.
Burkholder, J.M.; Wetzel, R.G. Epiphytic microalgae on natural substrata in a hardwater
lake: seasonal dynamics of community structure, biomass and ATP content. Arch.
Hydrobiol. Suppl. 83:1-56; 1989.
Carlton, R.G.; Wetzel, R.G. Distributions and fates of oxygen in periphyton communities.
Can. J. Bot. 65:1031-1037; 1987.
Carlton, R.G.; Wetzel, R.G. Phosphorus flux from lake sediments: effect of epipelic algal
photosynthesis. Limnol. Oceanogr. 33:562-570; 1988.
Carter, S.M. Herbivory by Donacia rufescens Lacordaire and Donacia cincticornis Newman on the white water-lily, Nymphaea odorata Aiton. M.Sc. thesis, Univ. Alabama,
Tuscaloosa; 1995.
7. Dynamics of Dissolved Organic Carbon
145
Cole, J.1.; Findlay, S.; Pace, M.L. Bacterial production in fresh and saltwater ecosystems: A
cross-system overview. Mar. Ecol. Pro gr. Ser. 43:1-10; 1988.
Coveney, M.F.; Wetzel, RG. Biomass, production, and specific growth rate of bacterioplankton and coupling to phytoplankton in an oligotrophic lake. Limnol. Oceanogr.
40:1187-1200; 1995.
del Giordio, P.A.; Peters, R.H. Patterns in planktonic P:R ratios in lakes: influence of alke
trophy and dissolved organic carbon. Limnol. Oceanogr. 39:772-787; 1994.
Dickerman, J.A.; Wetzel, RG. Clonal growth in Typha latifolia: population dynamics and
demography of the ramets. 1. Ecol. 73:535-552; 1985.
Fletcher, M. The physiological activity of bacteria attached to solid surfaces. Adv. Microb.
Physiol. 32:53-85; 1991.
Fletcher, M.; Floodgate, G.D. An electron-microscopic demonstration of an acidic polysaccharide involved in the adhesion of a marine bacterium to solid surfaces. 1. Gen.
Microbiol. 74:325-334; 1973.
Fletcher, M.; Marshall, K.C. Are solid surfaces of ecological significance to aquatic bacteria? Adv. Microb. Ecol. 6: 199-236; 1982.
Hessen, D.O. Dissolved organic carbon in a humic lake: effects on bacterial production and
respiration. Hydrobiologia 229:115-123; 1992.
Hossell, J.; Baker, J.H. Epiphytic bacteria of the freshwater plant Ranunculus penicillatus:
enumeration, distribution and identification. Arch. Hydrobiol. 86:322-337; 1979.
Hutchinson, G.E. A treatise on limnology. III. Limnological botany. 1\ew York: John Wiley
& Sons; 1975.
Jeppesen, E.; Sortkjaer, 0.; S0ndergaard, M.; Erlandsen, M. Impact of a trophic cascade on
heterotrophic bacterioplankton production in two shallow fish-manipulated lakes. Arch.
Hydrobiol. Beih. Ergebn. Limnol. 37:219-231; 1992.
Kim, B.; Wetzel, RG. The effect of dissolved humic substances on the alkaline phosphatase
and the growth of microalgae. Verh. Int. Verein. Limnol. 25: 122-128; 1993.
Kirchman, D.L.; Mazzella, L.; Mitchell, R; Alberte, RS. Bacterial epiphytes on Zostera
marina surfaces. BioI. Bull. 159:461-462; 1980.
Kirchman, D.L.; Mazzella, L.; Alberte, RS.; Mitchell, R. Epiphytic bacterial production on
Zostera marina. Mar. Ecol. Progr. Ser. 15: 117-123; 1984.
Laycock, R.A. The detrital food chain based on seaweeds. I. Bacteria associated with the
surface of Laminaria fronds. Mar. BioI. 25:223-231; 1974.
Lemke, M.1.; Churchill, P.F.; Wetzel, R.G. The effect of substrate and cell surface hydrophobicity on phosphate uptake in bacteria. Appl. Environ. Microbiol. 61:913-919;
1995.
Lock, M.A. River epilithon-a light and organic energy transducer. In: Lock, M.A;
Williams, D.O., eds. Perspectives in running water ecology. New York: Plenum Press;
1981:3-40.
Lock, M.A. The dynamics of dissolved and particulate organic material over the substratum
of water bodies. In: Wotton, R.S., ed. The biology of particles in aquatic systems. Boca
Raton, FL: CRC Press; 1990: 117-144.
Lock, M.A.: Wallace, R.R; Costerton, l.W.; Ventullo, R.M.; Charlton, S.E. River epilithon
(biofilm): toward a structural functional model. Oikos 42: 10--22; 1984.
Losee, R.E; Wetzel, RG. Littoral flow rates within and around submersed macrophyte
communities. Freshwat. BioI. 29:7-17; 1993.
Lyche, A; Andersen, T.; Christoffersen, K.; Hessen, D.O.; Hansen, P.H.B.; Klysner, A
Mesocosm tracer studies. 2. The fate of primary production and the role of consumers in
the pelagic carbon cycle of a mesotrophic lake. Limnol. Oceanogr. 41 :475-487; 1996.
Madsen, T.Y.; Warncke, E. Velocities of currents around and within submerged aquatic
vegetation. Arch. Hydrobiol. 97:389-394; 1983.
Mann, C.l.; Wetzel, RG. Dissolved organic carbon and its utilization in a riverine wetland
ecosystem. Biogeochemistry 31 :99-120; 1995.
146
R.G. Wetzel and M.
S~ndergaard
Mann, C.J.; Wetzel, R.G. Loading and utilization of dissolved organic carbon from emergent macrophytes. Aquat. Bot. 53:61-72; 1996.
Mazure, H.G.E; Field, J.G. Density and ecological importance of bacteria on kelp fronds in
an upwelling region. J. Exp. Mar. BioI. Ecol. 43:173-182; 1980.
Mickle, AM.; Wetzel, RG. Effectiveness of submersed angiosperm-epiphyte complexes on
exchange of nutrients and organic carbon in littoral systems. II. Dissolved organic
carbon. Aquat. Bot. 4:317-329; 1978.
Mickle, AM.; Wetzel, RG. Effectiveness of submersed angiosperm-epiphyte complexes on
exchange of nutrients and organic carbon in littoral systems. m. Refractory organic
carbon. Aquat. Bot. 6:339-355; 1979.
Moeller, RE.; Burkholder, J.M.; Wetzel, RG. Significance of sedimentary phosphorus to a
rooted submersed macrophytes (Najas flexilis (Willd.) Rostk. and Schmidt) and its algal
epiphytes. Aquat. Bot. 32:261-281; 1988.
Moran, M.A; Hodson, RE. Formation and bacterial utilization of dissolved organic carbon
derived from detrital lignocellulose. Lirnnol. Oceanogr. 34:1034-1047; 1989.
Moran, M.A; Zepp, RG. Role of photolysis in the formation of biologically labile compounds from dissolved organic matter. Limnol. Oceanogr. (in press).
Moriarty, DJ.W.; Boon, P.I.; Hansen, J.A; Hunt, W.G.; Poiner, I.R Microbial biomass and
productivity in seagrass beds. Geornicrobiol. J. 4:21-51; 1985.
Neely, RK.; Wetzel, RG. Simultaneous use of 14C and 3H to determine autotrophic
production and bacterial protein production in periphyton. Microb. Ecol. 30:227-237;
1995.
Newell, S.Y. Fungi and bacteria in or on leaves of eelgrass (Zostera marina L.) from
Chesapeake Bay. Appl. Environ. Microbiol. 41 :1219-1224; 1981.
Otsuki, A; Wetzel, R.G. Release of dissolved organic matter by autolysis of a submerged
macrophyte, Scirpus subterminalis. Limnol. Oceanogr. 19:842-845; 1974.
Peduzzi, P.; Herndl, G.J. Decomposition and significance of seagrass leaf litter (Cymodocea
nodosa) for the microbial foodweb in coastal waters (Gulf of Trieste, northern Adriatic
Sea). Mar. Ecol. Progr. Ser. 71:163-174; 1994.
Phillips, G.L.; Eminson, D.; Moss, B. A mechanism to account for macrophyte decline in
progressively eutrophicated fresh waters. Aquat. Bot. 4:103-126; 1978.
Raven, J.A. Energetics and transport in aquatic plants. New York: Alan R Liss, Inc.; 1984.
Riber, H.H.; Wetzel, RG. Boundary-layer and internal diffusion effects on phosphorus
fluxes in lake periphyton. Lirnnol. Oceanogr. 32:1181-1194; 1987.
Rich, P.H.; Wetzel, RG.; Thuy, N.V. Distribution, production and role of aquatic macrophytes in a southern Michigan marl lake. Freshwat. BioI. 1:3-21; 1971.
Roemer, S.C.; Hoagland, K.D.; Rosowski, J.R Development of a freshwater periphyton
community as influenced by diatom mucilages. Can. J. Bot. 62: 1799-1813; 1984.
Round, EE. The ecology of algae. Cambridge: Cambridge University Press; 1981.
Sand-Jensen, K.; Mebus, J.R. Fine-scale patterns of water velocity within macrophyte
patches in streams. Oikos 76:169-180; 1996.
Sculthorpe, C.D. The biology of aquatic vascular plants. New York: St. Martin's Press;
1967.
Sj/lndergaard, M. Heterotrophic utilization and decomposition of extracellular organic carbon (EOC) released by the aquatic angiosperm Littorella uniflora. Aquat. Bot. 16:59-73;
1983.
S~ndergaard, M. Extracellular organic carbon (EOC) in the genus Carpophyllum (Phaeophyceae): Diel release patterns and EOC lability. Mar. BioI. 104: 143-151; 1990.
Sj/lndergaard, M. Organic carbon pools in two Danish lakes: flow of carbon to bacterioplankton. Verh. Int. Verein. Limnol. 25:593-598; 1993.
S~ndergaard, M.; Sand-Jensen, K. Total autotrophic production in oligotrophic Lake Kalgaard, Denmark. Verh. Int. Verein. Lirnnol. 20:667-673; 1979.
7. Dynamics of Dissolved Organic Carbon
147
M.; Wetzel, R.G. Photorespiration and internal recycling of CO 2 in the
submersed angiosperm Scirpus subterminalis Torr. Can. J. Bot. 58:591-598; 1980.
Spencer, W.E.; Wetzel, R.G. Acclimation of photosynthesis and dark respiration of a
submersed angiosperm beneath the ice in a temperate lake. Plant Physiol. 101 :985-991;
1993.
Sutherland, LW. Biosynthesis and composition of gram-negative bacterial extracellular and
wall polysaccharides. Annu. Rev. Microbiol. 39:243-270; 1985.
Thomaz, S.M.; Wetzel, R.G. eHJleucine incorporation methodology to estimate epiphytic
bacterial biomass production. Microb. Ecol. 29:63-70; 1995.
Tranvik, L. Allochthonous dissolved organic matter as an energy source for pelagic bacteria
and the concept of the microbial loop. Hydrobiologia 29: 107-114; 1992.
van Loosdrecht, M.e.M.; Lyklema, J.; Norde, W.; Zehnder, A.J.B. Influence of interfaces
on microbial activity. Microbiol. Rev. 54:75-87; 1990.
Velji, M.L; Albright, L.J. Microscopic enumeration of attached marine bacteria of seawater,
marine sediment, fecal matter, and kelp blade samples following pyrophosphate and
ultrasound treatments. Can. J. Microbiol. 32:121-126; 1986.
Westlake, D.E Comparisons of plant productivity. BioI. Rev. 38:385--425; 1963.
Wetzel, R.G. Limnology. 2nd Ed. Philadelphia: Saunders College Publishing; 1983a.
Wetzel, R.G. Attached algal-substrata interactions: fact or myth, and when and how? In:
Wetzel, R.G., ed. Periphyton in freshwater ecosystems. The Hague: Dr. W. Junk Publishers; 1983b:207-2l5.
Wetzel, R.G. Land-water interfaces: metabolic and limnological regulators. Verh. Int.
Verein. Limnol. 24:6-24; 1990.
Wetzel, R.G. Extracellular enzymatic interactions in aquatic ecosystems: storage, redistribution, and interspecific communication. In: Chrost, R.I., ed. Microbial enzymes in
aquatic environments. New York: Springer-Verlag; 1991:6-28.
Wetzel, R.G. Gradient-dominated ecosystems: sources and regulatory functions of dissolved organic matter in freshwater ecosystems. Hydrobiologia 229:181-198; 1992.
Wetzel, R.G. Microcommunities and microgradients: linking nutrient regeneration, microbial mutualism, and high sustained aquatic primary production. Netherlands J. Aquat.
Ecol. 27:3-9; 1993a.
Wetzel, R.G. Humic compounds from wetlands: complexation, inactivation, and reactivation of surface-bound and extracellular enzymes. Verh. Int. Verein. Limnol. 25: 122-128;
1993b.
Wetzel, R.G. Death, detritus, and energy flow in aquatic ecosystems. Freshwat. BioI.
33:83-89; 1995.
Wetzel, R.G. Benthic algae and nutrient cycling in lentic freshwater ecosystems. In: Stevenson, R.J.; Bothwell, M.L.; Lowe, R.L., eds. Algal ecology: freshwater benthic ecosystems. New York: Academic Press; 1996:641-667.
Wetzel, R.G.; Allen, H.L. Functions and interactions of dissolved organic matter and the
littoral wne in lake metabolism and eutrophication. In: Kajak, Z.; Hillbricht-Illkowska,
A., eds. Productivity problems of fresh waters. Warsaw: PAN Publishers; 1970:333347.
Wetzel, R.G.; Howe, M.I. Population dynamics and seasonal growth and biomass patterns
of an emergent rush (Juncus effusus L.) in a subtemperate Alabama wetland. Aquat. Bot.
(in press).
Wetzel, R.G.; Rich, P.H.; Miller, M.e.; Allen, H.L. Metabolism of dissolved and particulate
detrital carbon in a temperate hard-water lake. Mem. 1st. Ital. Idrobiol. 29(suppl.): 185243; 1972.
Wetzel, R.G.; Brammer, E.S.; Forsberg, e. Photosynthesis of submersed macrophytes in
acidified lakes. I. Carbon fluxes and recycling of CO 2 in Juncus bulbosus L. Aquat. Bot.
19:329-342; 1984.
S~ndergaard,
148
R.G. Wetzel and M. S¢ndergaard
Wetzel, R.G.; Hatcher, P.G.; Bianchi, T.S. Natural photolysis by ultraviolet irradiance of
recalcitrant dissolved organic matter to simple substrates for rapid bacterial metabolism.
Limnol. Oceanogr. 40: 1369-1380; 1995.
Wetzel, R.G.; Ward, A.K.; Stock, M. Effects of natural dissolved organic matter on
mucilaginous matrices of biofilm communities. Arch. Hydrobiol. 140: I-II; 1997.
8.
Impact of Herbivory on Plant Standing Crop:
Comparisons Among Biomes, Between
Vascular and Nonvascular Plants, and Among
Freshwater Herbivore Taxa
David M. Lodge, Greg Cronin, Ellen van Dank, and
Adrienne J. Froelich
Introduction
Two contradictory traditions exist regarding the impact of herbivores on the
ecology and evolution of plants. For ecologists studying terrestrial ecosystems, the
interaction between plants and their consumers has been a focal point for research
in recent decades. Herbivores are widely regarded as an important determinant of
plant abundance and species composition and as an important selective force in
the evolution of terrestrial plant traits (Rhoades, 1985; Herms and Mattson, 1992;
Rosenthal and Berenbaum, 1992). Similarly, the abundance of many marine plants
is often reduced by herbivores, and many seaweed traits are thought to have
evolved in response to herbivory (Lubchenco and Gaines, 1981; Gaines and
Lubchenco, 1982; Estes and Steinberg, 1988; Hay, 1991). By contrast, for decades
the paradigm in limnology has been that live freshwater macrophytes are too tough
for the mouthparts of aquatic herbivores, are of low nutritional quality, and are
rarely consumed by herbivores (Lodge, 1991; Newman, 1991).
The idea that freshwater herbivores are unimportant has recently come under
more careful scrutiny. Data contradict the assumption of low nitrogen content
(Duarte, 1992) and suggest that live macrophytes may be consumed far more than
is often appreciated (Lodge, 1991). Whereas many past freshwater studies documented large reductions of periphytic algae by herbivores (Feminella and Hawkins, 1995; Lamberti, 1996; Steinman, 1996), an increasing number of direct tests
149
150
D.M. Lodge et al.
of the impact of freshwater herbivores on macrophyte abundance are now being
conducted.
If freshwater herbivores reduce aquatic macrophyte biomass or change the
species composition of plants in lakes, they could modify the many important
biotic, abiotic, and biogeochemical processes in which macrophytes are involved
(Carpenter and Lodge, 1986; many chapters in this volume). Given this potentially important role of freshwater herbivores in community and ecosystem
function, it becomes essential to better understand freshwater macrophyteherbi vore interactions.
In this chapter, we use recent studies to evaluate the impact of herbivores on
freshwater macrophytes. First, we compare the magnitude of impact of herbivores
on plant standing crop in freshwater, marine, and terrestrial habitats. If the magnitude of herbivore impact on freshwater macrophytes is at least as great as that in
other biomes, then there is no a priori reason to believe its evolutionary or
ecological importance is less than that of terrestrial herbivores.
Second, for freshwater and marine habitats, we compare the herbivore impact
on nonvascular with that on vascular plants. Interbiome comparisons could be
partly confounded by the different taxonomic affinities of the plants common to
different biomes. Marine studies focus on macroalgae (because few marine vascular species exist), terrestrial studies focus on vascular plants, and freshwater
studies have historically focused on microalgae (but are increasingly addressing
macroalgae and vascular macrophytes). Perhaps algae are generally more susceptible to herbivory than vascular macrophytes because most algae contain less
structural materials (e.g., cellulose, lignin, cuticles), which render some vascular
plants very tough and/or indigestible (Grubb, 1986; Rosenthal and Berenbaum,
1992), and because the lack of vascular tissues mean algae cannot efficiently
transport defensive compounds (Cronin and Hay, 1996b). Or the typically higher
growth rates of algae may simply allow them to better tolerate herbivory (Lubchenco and Gaines, 1981). Our comparison of herbivore impact will test whether
algae are more reduced by herbivores than vascular plants.
Third, narrowing our focus to the freshwater habitat, we compare the impact on
both algae and macrophytes of different taxa of herbivores. Most freshwater
studies of herbivory have focused on insects and snails. It could be that other
groups of herbivores, including crustaceans, fishes, turtles, mammals, and aquatic
birds, have an equal or larger impact on freshwater plants that has simply not been
measured because the traditional training of limnologists has not included work
with these taxa.
Fourth, because our analysis suggests that some of these other taxa may, in fact,
often reduce macrophyte standing stock, we briefly review literature suggesting
that the spatial and temporal dynamics of some of these taxa are complex and
require much more thorough study by lirnnologists. We address the hypothesis that
the abundance and impact of mammalian and avian herbivores have been lower
during the past century than during most of the evolutionary history of freshwater
plants because of anthropogenic reductions in their populations. We also present
data that suggest this trend has reversed, at least in North America and northern
8. Impact of Herbivory on Plant Standing Crop
151
Europe. Thus, we might expect herbivory by mammals and birds to be increasingIy important.
Fifth, we use the literature to assess the degree of diet specialization of different
herbivore taxa. Available data suggest that many of the least studied taxa (which
may also be increasing in importance) are generalists (i.e., consume many macrophyte species in several plant families). To understand and predict their impact on
lake macrophyte communities, we propose a conceptual model of feeding selectivity by generalist herbivores.
Sixth and finally, we evaluate how the impact of herbivores on macrophyte
abundance and species composition may affect the shifts of lakes between the
clear water and turbid states.
Does Herbivore Impact Differ Among Terrestrial, Marine, and
Freshwater Biomes, and Between Vascular and Nonvascular Plants?
Using previously published primary and secondary literature sources, we compared the reduction in standing crop of vascular and nonvascular plants (excluding
phytoplankton) caused by herbivores in terrestrial, marine, and freshwater habitats
(Fig. 8.1). With respect to the herbivores, we excluded studies on livestock and on
herbivores introduced specifically to control plants. With respect to methods, we
used both experimental studies (in which herbivores were directly excluded and/or
included with cages) and comparisons of otherwise similar unmanipulated sites
with and without herbivores. From all studies, we calculated the percentage
reduction of plant abundance (A) as A = [(A-berbivore - A+herbivore)/A-berbivore] 100. We
used whatever index of abundance that the original author used (e.g., biomass,
numbers, leaf area damaged).
This or similar indices have been used to estimate grazer consumption and the
impact of herbivores on plant productivity (Cyr and Pace, 1993; Cattaneo and
Mousseau, 1995), but such uses can lead to errors of large magnitude because of
the timing or duration of experiments, density-dependent plant growth, and uncontrolled herbivore feedbacks (Jacobsen and Sand-Jensen, 1994; Mitchell and Wass,
1996a; Wass and Mitchell, this volume, Chapter 18; Mitchell and Perrow, this
volume, Chapter 9). Thus, with our comparisons, we do not imply anything about
consumption rates of herbivores or about the impact of herbivores on plant
production. Rather, we use it solely as an index of the differences in plant standing
stock that result from the entangled set of complex interactions that differ in the
presence and absence of herbivores. Yet even this application requires considerable caution in interpretation because plant abundance A is often very dependent
on the timing and duration of a study relative to the growth cycle of the plants
(Mitchell and Wass, 1996a; Wass and Mitchell, this volume, Chapter 18; Mitchell
and Perrow, this volume, Chapter 9). Nevertheless, A is the statistic that can be
calculated from the greatest number of studies, and few studies provide the time
course of data necessary to evaluate the time dependence of A. In our literature
survey, the spatial and temporal scale of studies ranged widely and no doubt
152
D.M. Lodge et a!.
_-Vascular
~
r:::::::=:::J
100
~
"-
f
~
,
Q)
-Nonvascular
80
.cOD
~'
~
§
.... 60
g
+I
::::
Q)
::::::'§
.:2 E
t)
;::l
~
40
20
O~--~--------L-----L------
Terrestrial
(75,0)
Marine
(17,52)
Freshwater
(70,32)
Figure 8.1. Mean and range of percentage reduction of standing crop of vascular and
nonvascular plants (excluding phytoplankton) in terrestrial, marine, and freshwater habitats.
Sample numbers (numbers of separate experiments or comparisons) are in parentheses
below the habitat name (vascular, nonvascular). The few negative values were not plotted
but are listed here: -2, -20, -38 for terrestrial; -5, -5, -50 for marine; -24, -47 for
freshwater. (Data for freshwater habitat from Abrahamsson, 1966; Dean, 1969; Pelkian et
aI., 1971; Soska, 1975; Urban, 1975; Anderson and Low, 1976; Jupp and Spence, 1977;
Verhoeven, 1978; Kiorboe, 1980; van der Velde, 1982; Prejs, 1984; Carter and Rybicki,
1985; Smith and Kadlec, 1985; Wallace and O'Hop, 1985; Hansson et a!., 1987; Lodge and
Lorman, 1987; Scott and Haskins, 1987; Korshgen et aI., 1988; Painter and McCabe, 1988;
Esler, 1989; Feminella and Resh, 1989; Crutchfield et a!., 1992; Hart, 1992; Jacobsen and
Sand-Jensen, 1992, 1994, 1995; Julien et a!., 1992; Maceina et aI., 1992; Martin et aI., 1992;
Matthews and Reynolds, 1992; Shaffer et aI., 1992; Underwood et al., 1992; Wootton and
Oemke, 1992; Samelle et aI., 1993; Urbanc and Blejec, 1993; Bronmark, 1994; Chikwenhere, 1994; Conover and Kania, 1994; Creed, 1994; Hoyer and Canfield, 1994; Komij6w,
1994, 1996; Lodge et aI., 1994; Schulten et aI., 1994; Taylor et aI., 1994; van Donk et a!.,
1994; Vermaat, 1994; Feminella and Hawkins, 1995; Richardson et aI., 1995; Taylor and
Grace, 1995; van Donk and Gulati, 1995; S~mdergaard et aI., 1996. Data for marine habitat
from Castenholz, 1961; Randall, 1961, 1965; Sammarco et aI., 1974; Wanders, 1977;
Ogden et aI., 1979; Vance, 1979; Duggins, 1980; Sammarco, 1980, 1982; Carpenter, 1981,
1986; Hatcher, 1981; Hay, 1981, 1984; Gaines and Lubchenco, 1982; Hatcher and Larkum,
1983; Hay and Goertemiller, 1983; Hay et aI., 1983; Himmelman et aI., 1983; Bertness,
1984; Cargill and Jefferies, 1984,1986; Hay and Taylor, 1985; Bazely and Jeffries, 1986;
Lewis, 1986; Fletcher, 1987; Foster, 1987; Lewis et aI., 1987; Russ, 1987; van Tamelan,
1987; Morrison, 1988; Santelices and Martinez, 1988; Buschmann, 1990; Keats et aI.,
1990; Geller, 1991; Valentine and Heck, 1991; Jones, 1992; Polunin and Klumpp, 1992;
Andrew and Underwood, 1993; Benedetti-Cecchi and Cinelli, 1993; Mitchell et aI., 1994;
Estes and Duggins, 1995; Heck and Valentine, 1995; Prince, 1995; Steinberg et aI., 1995;
Miller and Hay, 1996. Data on terrestrial habitats from Witkowski, 1980; Yden berg and
Prins, 1981; Coppock et al., 1983; McNaughton, 1985; Hughes et aI., 1987; Huntley, 1987;
Owen-Smith and Cooper. 1987; Risley and Crossley, 1988; Whicker and Detling, 1988;
Coley and Aide. 1991; Marquis and Whelan, 1994; Bonser and Reader, 1995; Hulme, 1996;
Schreiner et aI., 1996.)
8. Impact of Herbivory on Plant Standing Crop
153
affected results of individual studies. We made no attempt to account for the scale
at which studies were performed. We suspect that most investigators select the
experimental duration that maximizes the differences in standing stock between
enclosures and exclosures, thus maximizing our A. For example, in temperate zone
habitats where winter foliar standing stock is typically near zero regardless of
herbivores (A = 0), investigators end their experiment before seasonal senescence
of plants begins so that at least the potential exists for A to exceed zero. Thus
although most results reported in the literature may produce impact indices
that are more maximal than minimal, we are not aware of any systematic
methodological differences that would bias our comparisons across habitat
types and plant types.
For estimating some ecosystem impacts (e.g., nutrient cycling), knowledge of
plant production and how it is affected by herbivory might be important (Mitchell
and Wass, 1996a), but plant standing stock per se is an important determinant in
community and ecosystem interactions involving, for example, predation refuge,
microclimate effects, and provision of surface areas for epiphytic organisms. Thus,
our focus on differences in standing stock during the growing season has great
applicability to these latter interactions.
For the terrestrial habitat, we relied primarily on secondary sources (e.g., Coley
and Aide, 1991, on tree leaves), but added some recent primary literature on
terrestrial vegetation types not well represented in the secondary sources (e.g.,
studies by McNaughton and colleagues on grasslands). Terrestrial nonvascular
plants are not included because few relevant data exist. For the marine habitat, we
relied completely on primary sources because no reviews have been published.
For vascular freshwater plants, we added more recent sources to those found in
Lodge (1991) and Newman (1991). For nonvascular freshwater plants, we relied
heavily on the review by Feminella and Hawkins (1995). Complete information on
sources is provided (see Fig. 8.1 legend). Using the results of our literature survey,
we make qualitative comparisons only, because in some cases we did not have
access to the original data (e.g., from review articles), which would be necessary
to calculate anything other than a mean and range. Statistical comparisons were
not possible for these data.
Results suggest that herbivores in all three habitats may often reduce plant
abundance substantially (Fig 8.1). Herbivore impact on nonvascular plants overlaps broadly with that on vascular plants, but in freshwater, mean impact on
nonvascular plants is greater. Freshwater nonvascular plants might be more susceptible to a wide variety of herbivores because oftheir smaller size, lack of tough
structural material, and possibly lower levels of defensive compounds. Herbivory
on nonvascular plants in marine habitats is comparable with that in freshwater
habitats, but to some extent the plant groups being compared are different. Most
marine studies focus on rnacroalgal seaweeds (predominantly Phaeophyta, Rhodophyta, and Chlorophyta), for which there are almost no similar-sized counterparts
in freshwater habitats. Most freshwater studies focus on microalgae (predominantly Chlorophyta, Chrysophyta, and Cyanobacteria), which also occur in marine
habitats (including on the surfaces of macroalgae) but have attracted less study in
154
D.M. Lodge et al.
marine habitats. The implications of these interhabitat differences in algal taxa and
anatomy for the plant-herbivore interaction are largely unknown.
For vascular plants, herbivore impact is similar in terrestrial and freshwater
habitats and considerably higher in marine habitats. We suspect that the means for
both terrestrial and freshwater habitats are biased low because both terrestrial
(Dirzo and Miranda, 1991) and freshwater (see next section) plant-herbivore
ecologists have been preoccupied with insects, which tend to have lower impact
than many other herbivores on vascular plants. However, even the current data
(Fig. 8.1) suggest that freshwater herbivores reduce macrophyte abundance substantially, thus confirming earlier suggestions (Lodge, 1991) that herbivory is at
least as important in freshwater as it is in terrestrial habitats. It is therefore
reasonable to hypothesize that freshwater herbivores may often exert substantial
selection pressure on plant characteristics and alter the role of macrophytes in
aquatic ecosystem function. It is important to understand in more detail what sort
of freshwater herbivores most reduce macrophytes and what role herbivory plays
in altering the ecosystem impact of macrophytes. For the remainder of this chapter, we narrow our focus to the plant-herbivore interaction in freshwaters.
Does Herbivore Impact Differ Among Freshwater Herbivore Taxa?
To compare the impact of different taxa of freshwater herbivores, we replotted the
freshwater data from Figure 8.1 by herbivore type (Fig. 8.2). The absence of
studies addressing the impact on periphyton of herbivores with larger body sizemammals and aquatic birds-reflects the widely held belief (apparently supported
by abundant observations) that these herbivores rarely intentionally consume
periphyton (although counter examples exist, such as black swan studies in
Mitchell et aI., 1988; Mitchell and Perrow, this volume, Chapter 9), but may often
consume epiphytic periphyton incidentally with macrophytes. Intermediate-sized
herbivores-fishes and crayfishes-reduce macrophytes and periphyton similarly,
with the literature suggesting that the periphyton consumed by these species is
primarily filamentous macroalgae that may be intentionally ingested by the herbivores. By contrast, the smallest herbivores-snails and insects-have little
(insects) or no impact (snails) on macrophytes but do substantially reduce periphyton, especially microalgae (Fig. 8.2).
For periphyton, the largest mean percentage reductions were by fishes and crayfishes. In some U.S. midwestern streams, herbivorous fishes (Cyprinidae), whose
distribution and abundance among stream pools is determined by piscivorous fishes
(Centrarchidae), determine whether pools are green with fIlamentous algae or grazed
bare (Matthews et al., 1987; Power et al., 1988). Similarly, crayfishes can exert control
over the abundance of stream Cladophora (Hart, 1992; Creed, 1994). By contrast, the
most studied herbivore taxa, snails and insects (see sample numbers on Fig. 8.2), had
lower but still large mean reductions of periphyton.
For vascular plants, it is clear from comparing the number of studies on
different taxa (Fig. 8.2) that past bias in studying insects to the exclusion of other
155
8. Impact of Herbivory on Plant Standing Crop
100
~
=<
0 ......
g<
rzt...t
~'-'
..
0
0
40
:::::::. 20
0
•
•••
80
"'::: of 60
"8'
.-Vascular
O-Nonvascular
•
• i
-I-
•
Mammal
(7,0)
....
••
•
Bird
(16,0)
-AI
;
•
•
; ~;
i_
•
•• &
0
•
an
•
0
0
Q>
(]
cg
•
0
=
°D
cB
c9
•
Fish
(7,10)
@
8
8
Crayfish
(5,3)
Snail
(0,38)
•
0
•
"
:K
Insect
(32,24)
Figure 8.2. Data points and means (bars) for the percentage reduction of standing crop of
freshwater vascular and nonvascular plants (excluding phytoplankton) by different taxa of
herbivores. Herbivore taxa are listed in rough order of declining body size from left to right.
Large dots are mean values from several studies summarized in Feminella and Hawkins
(1995). Small dots represent one experiment or comparison. Sheldon (1987), which seemed
to show large reductions of macrophytes by snails, was excluded because later communication suggested that snails had not consumed the macrophytes and probably were not
responsible for the macrophyte reductions (Bronmark, 1990; Sheldon, 1990). Sample
numbers (numbers of separate experiments or comparisons) are given below the name of
the herbivore taxon (vascular, nonvascular). Values of -24 and -47 were not plotted for
birds. (Data on birds from Anderson and Low, 1976; Jupp and Spence, 1977; Verhoeven,
1978; Kiorboe, 1980; Carter and Rybicki, 1985; Korshgen et al., 1988; Esler, 1989; Urbanc
and Biefec, 1993; Conover and Kania, 1994; Hoyer and Canfield, 1994; Schutten et al.,
1994; van Donk et aI., 1994; van Donk and Gulati, 1995; Sjljndergaard et al., 1996. Data on
crayfish from Abrahamsson, 1966; Dean, 1969; Lodge and Lorman, 1987; Feminella and
Resh, 1989; Hart, 1992; Matthews and Reynolds, 1992; Lodge et aI., 1994. Data on fish
from Prejs, 1984; Hansson et aI., 1987; Crutchfield et al., 1992; Maceina et aI., 1992;
Wootton and Oemke, 1992; Urbanc and Blejec, 1993; Creed, 1994; Feminella and Hawkins, 1995; Richardson et al., 1995. Data on insects from Soska, 1975; Urban, 1975; van der
Velde, 1982; Wallace and O'Hop, 1985; Scott and Haskins, 1987; Painter and McCabe,
1988; Jacobsen and Sand-Jensen, 1992, 1994, 1995; Underwood etal., 1992; Chikwenhere,
1994; Komij6w, 1994, 1996; Feminella and Hawkins, 1995. Data on mammals from
Pelikan et aI., 1971; Smith and Kadlec, 1985; Julien et aI., 1992; Shaffer et aI., 1992; Taylor
et al., 1994; Taylor and Grace, 1995. Data on snails from Martin et aI., 1992; Samelle et al.,
1993; Bronmark, 1994; Vermaat, 1994.)
herbivores has distorted our view of the role of herbivory and contributed to what
is probably an artifactually low mean for the freshwater habitat (Fig. 8.1). About
48% of all the studies on herbivory of macrophytes have been done on insects. Yet
they have the smallest impact on macrophytes of any herbivore group (Fig. 8.2).
The largest reductions of macrophyte biomass were caused by the lesser-studied
herbivores (Le., crayfishes, fishes, and aquatic birds). Mammals, primarily muskrat
156
D.M. Lodge et al.
(Ondatra zibethicus) and nutria (Myocastor cOYpus), typically caused reductions
of macrophyte standing crop intermediate between insects and other vertebrates
(Fig. 8.2).
An additional reason for caution in interpreting these patterns is that most
experiments targeted one species or guild of herbivores. In many habitats, other
herbivores were probably present, but their impact and any direct or indirect effect
on their impact by the manipulation of the target herbivore were usually not
quantified.
Some important biogeographical differences in the occurrences of macrophyteeating herbivores are obscured by these data summaries. For example, there are no
fishes native to North America for which macrophytes are an important dietary
component. Macrophyte-eating fishes are, however, common in Europe (e.g.,
Rutilus rutilus, Scardinius erythrophthalmus, Cyprinus carpio) and contribute to
reductions in macrophytes that may destabilize the clearwater state of shallow
northern European lakes (van Donk and OUe, 1996). Macrophyte-eating fishes are
also common in Asia (e.g., Ctenopharygodon idella), Africa, and South America
(many species of Cichlidae and Characidae). Such biogeographical patterns interact with other natural and anthropogenic spatial and temporal patterns in herbivore
impact that operate on much shorter time scales. In the next section, we narrow our
focus further to vascular plants only. We consider how short-term spatial and
temporal variation in herbivore abundance may differ among taxa and how that
variation may affect the plant-herbivore interaction.
Spatial and Temporal Dynamics of Herbivore Abundance
One difficulty in interpreting the patterns of herbivore impact from the literature
(Figs. 8.1 and 8.2) is knowing what spatial and temporal scales the documented
reductions of macrophytes by herbivores represent. Evidence suggests that in the
absence of major anthropogenic changes in the lake environment (e.g., addition or removal of herbivores or exotic plant species, eutrophication, acidification), macrophyte abundance and species composition are remarkably constant
over decades, at least in oligotrophic-mesotrophic lakes (Carpenter and Titus,
1984; Lodge et aI., 1989). Thus impact of changes in herbivory can be distinguished from background variation in macrophyte abundance. It then becomes
essential to consider the temporal and spatial patterns of variation in the abundance and impact of herbivores.
Invertebrates, Fishes, and Turtles
The natural abundance of many taxa of herbivores and their impact on macrophytes may be relatively similar over wide geographical regions and relatively
constant from year to year. Natural interannual variation certainly exists for many
invertebrates and is well documented for many fishes. However, natural disappearance or appearance of herbivore species on an interannual time scale is
8. Impact of Herbivory on Plant Standing Crop
157
probably rare, such that macrophytes are relatively constantly subject to herbivory
exerted by many invertebrates, fishes, and turtles (Clark and Gibbons, 1969;
Carter and Rybicki, 1985).
By contrast, anthropogenic interannual variation induced by harvesting, other
management practices, and eutrophication can be large and is especially well
documented for fishes. For example, eutrophication leads to increases in zooplanktivorous and benthi vorous fishes in northern Europe (Persson et aI., 1991),
with consequences for water clarity and macrophyte abundance that are beginning
to be understood from subsequent removal of zooplanktivorous fishes (e.g., Rutilus rutilus removal in biomanipulation).
Other anthropogenically driven changes important for macrophyte abundance
and lake management involve the spread of exotic species of herbivorous invertebrates and fishes. Probably the most important North American examples include three cyprinid fishes: the goldfish (Carassius auratus), introduced in the late
1600s; the common carp (Cyprinus carpio), introduced in 1831; and the grass carp
(Ctenopharygodon idella), introduced in the early 1960s. By the late 19th century,
the common carp was already extremely abundant throughout eastern North
America (Laird and Page, 1996), and large impacts on macrophytes, invertebrates,
and other fishes are well documented (Taylor et aI., 1984). Grass carp were
imported specifically to reduce nuisance macrophytes, which they have done very
effectively (Shireman, 1984).
The potential role of crayfishes as herbivores has been highlighted by reductions in macrophytes as crayfishes have been introduced outside their native range
(Lodge and Hill, 1994). The best studied example is the rusty crayfish (Orconectes
rusticus), a native of Indiana, that has been widely introduced elsewhere in eastern
North America. Its establishment has caused large declines in submerged macrophytes and snails in lakes (Lodge et aI., 1994). Introductions of exotics are
increasing globally and include both aquatic plants and herbivores; thus, new
plant-herbivore interactions can be expected in lakes (Lodge et aI., in press),
with potentially large consequences for macrophyte abundance and ecosystem
function.
Mammals
The reduction or elimination by humans of large mammalian herbivores in terrestrial ecosystems has been well documented (Wilson, 1992). Although less
studied, the pattern has been similar for freshwater mammalian herbivores that
were dramatically reduced in earlier centuries by hunting, trapping, and other
human activities (e.g., beaver [Castor canadensis], manatees [Trichechus manatus]) (Henderson, 1960; Campbell and Irvine, 1977; Whitaker, 1980; Jones and
Birney, 1988; Lacki et aI., 1990; Wilsey and Chabreck, 1991; Shaffer et aI., 1992;
Doucet and Fryxell, 1993). Other freshwater mammalian herbivores have, however, increased their geographical range, even as populations have fluctuated in
response to changing hunting and fur-trading pressure. Introductions of muskrat
(Ondatra zibethicus) into Europe (Hengeveld, 1989) and of nutria (Myocastor
158
D.M. Lodge et al.
coypus) into North America (Wilsey and Chabreck, 1991; Shaffer et aL, 1992) are
striking examples. Limnologists need to consider more carefully the impact of
such herbivores on lake macrophytes as their ranges and abundance change.
Because many of the most dramatic changes in abundance of these herbivores
happened so long ago, it is difficult to know what the impact of their removal
or addition was. We should not, however, simply assume that the impact was
negligible.
Aquatic Birds
Both the evolution and past ecology of macrophytes may have been strongly
affected by waterfowL Because limnologists have not often considered this possibility, we devote disproportionate space to waterfowl (relative to better studied
freshwater herbivores) in this chapter (also see Mitchell and Perrow, this volume,
Chapter 9). North American fossils (Bickart, 1990), northern European rock
etchings (Maringer and Bandi, 1953), and other evidence indicate that waterfowl,
including swans, which can consume up to 9 kg per capita of aquatic vegetation
daily (Bellrose, 1980), have been feeding from northern hemisphere lakes for
about a million years. However, since the 19th century, abundance of aquatic birds
has been very low relative to prehistorical levels.
Umegulated hunting in North America before 1916, the loss of 53% of all
wetlands in the continental United States (Dahl, 1990), and the degradation of
remaining wetlands caused a significant decline in waterfowl populations (Bellrose, 1980; Baldassare and Bolen, 1994). Although population estimates for most
waterfowl species do not exist for the time before European colonization, data for
trumpeter swan (Cygnus buccinator) may be representative of the trend for other
aquatic birds. Before European colonization, there were probably 100,000 trumpeter swans in the Great Lakes region of North America; yet, by 1900 there were
none (Gillette and Shea, 1995). Clearly, not only the absolute numbers of swans
(and probably other waterfowl) declined precipitously, but their numbers per unit
area of lake and wetland habitat have also been very low in the past century; the
decline in waterfowl numbers has been substantially greater than the decline in
wetland habitat area. Therefore, studies of herbivory on macrophytes and the
general impressions of limnologists about the importance of herbivory have been
shaped during what has probably been the period of lowest densities of herbivorous aquatic birds in the past many thousands of years.
Although quality and quantity of habitat continue to decline in both North
America and Europe, population trends for most herbivorous waterfowl have
reversed in recent decades in North America (Fig. 8.3) and Europe. Over the past
40 years in North America, diving duck populations have remained roughly
constant, but dabbling ducks have increased about 35%, geese about 225%, and
swans about 80% (Fig. 8.3). Over the past 25 years in northern and central Europe,
popUlations of three of the four most important herbivorous species have also
increased. While coot (Fulica atra) numbers have been relatively constant or
declining slightly, mute swans (Cygnus olor) have increased about 15%, Bewick's
8. Impact of Herbivory on Plant Standing Crop
159
~~------------------------------,
7000
6000
5000
!crJ
z
I<l
N
crJ
Z
o
~
~
4000
3000
2000
1000
2504-------------------------------;
200
150
100
.,.,
.,.,
50~--_r--_r--_r--,_--~--._--._--~
a"I
.-<
YEAR
Figure 8.3. Abundance of herbivorous aquatic birds in North America during the latter half
of the 20th century. Data have been pooled across species with similar eating habits:
dabblers (Anas americana, A. strepera), which typically eat seeds and other parts of macrophytes in water <0.5 m deep; divers (Aythya americana, A. valisinaia), which commonly
eat macrophytes at 1-2-m depth; geese (Branta canadensis, Chen caerulescens), which eat
terrestrial plants and/or emergent macrophytes; and swans (Cygnus buccinator, C. columbianus, C. alar), which feed on emergent and submerged macrophytes at water depths down
to 1 m. Data for coots (Fulica americana), the only major herbivorous aquatic bird not
included, were unavailable. All groups include both aboveground and belowground feeders.
Data are estimates of continentwide populations except for C. alar (for which only Atlantic
Flyway data are available). Data for geese before 1970 are not included because census
methods differed from later data. A few missing data points for some swan species were
estimated by interpolation. (Data from Caithamer and Dubovsky, 1995, for all ducks;
Caithamer and Dubovsky, 1995 (1970-1995) and courtesy of U.S. Fish and Wildlife Service
(1955-1970) for geese and tundra swans; and Allin et aI., 1987; Gillette and Shea, 1995;
Allin, personal communication, for other swans.)
160
D.M. Lodge et al.
swans (c. bewickii) have increased about 30%, and red-crested pochards (Netta
rufina) have increased about 15,000% (Rose, 1994).
The impacts that these higher densities of waterfowl may have on macrophytes
in wetlands, shallow lakes, and estuaries have been best studied for snow geese
(Chen caerulescens) in North America. Higher populations of snow goose have
resulted in reductions in a variety of terrestrial and emergent plant variables:
cover, biodiversity, aboveground standing crop, productivity, nitrogen content,
and belowground biomass (Smith and Odum, 1981; Cargill and Jefferies, 1984;
Bazely and Jefferies, 1986; Giroux and Bedard, 1987; Rockwell et at, 1996).
Snow geese predominantly affect emergent and wetland plants, but large reductions of submerged macrophytes and macroalgae by coots in Europe (van Donk et
al., 1994; Sjilndergaard et al., 1996; van Donk and Otte, 1996; van Donk, this
volume, Chapter 19), black swans in New Zealand (Mitchell et at, 1988; Mitchell
and Wass, 1996b), and diving ducks in Texas (Mitchell et aI., 1994) have also been
well documented in recent years (see Fig. 8.2). Clearly, lirnnologists need to
examine more carefully when, where, and by how much aquatic birds reduce
macrophyte abundance. Because of the strong seasonal migratory patterns of
many aquatic birds and the longer-term population trends, judging how widely
results of one study may apply is difficult.
Spatial and Temporal Patterns ofAquatic Bird Abundance
On large spatial scales (lake-to-Iake) and long time scales (year-to-year), a positive
relationship exists between abundance of herbivorous waterfowl and macrophyte
abundance (McAtee, 1911; Wilson and Atkinson, 1995; Mitchell and Wass, 1996b).
Herbivorous waterfowl choose lakes with higher macrophyte biomass and preferred species composition (Lovvorn, 1989; Squires, 1991; Baldassare and Bolen,
1994). The high mobility of aquatic birds relative to their resources makes this
positive relationship between consumer and resource unsurprising (Sih, 1984) and
suggests that birds may usually move before eliminating plants (Reinecke et at,
1989). However, ample evidence suggests that birds often reduce macrophyte
abundance during the periods they inhabit a lake. What remains almost untested
(with the exception of the work of Jefferies and colleagues cited above), is the
long-term impact of seasonal plant depletion, especially if the consumption is not
constant year to year. In this section, we can only begin to suggest the issues that
lirnnologists need to address before any general conclusions on the impact of
waterfowl on macrophytes are reached (see Mitchell and Perrow, this volume,
Chapter 9).
Waterfowl that are territorial on the breeding grounds (e.g., most dabblers,
divers, and swans except black swans) may have a low impact on macrophytes
because bird population densities are low. In North America, maximum densities
of adult breeding birds are about 5.4/ha and 1.6/ha, respectively, for dabbling and
diving ducks (Kantrud and Stewart, 1977), whereas the ranges of densities are
about 0.03-O.07/ha for trumpeter swan (Cygnus buccinator; Banko, 1960), and
0.05-O.1/ha for mute swan (c. olor; Wood and Gelston, 1972). In addition,
8. Impact of Herbivory on Plant Standing Crop
161
breeding females and young are primarily carnivorous (Baldassare and Bolen,
1994). By contrast, colonially nesting species, like the snow goose, may have
dramatic local effects on their resources and habitat---creating barren "eat-outs"during the breeding season (Rockwell et aI., 1996). Coots also may reduce macrophyte standing crop during the growing season, but like many other aquatic birds,
the most obvious impacts appear during autumn and winter aggregations (S0ndergaard et aI., 1996; van Donk and Oue, 1996, van Donk, this volume, Chapter 19).
During migrations and on the wintering grounds, almost all species are gregarious. These large and diverse assemblages of waterfowl could have a large
impact on both aboveground biomass and overwintering structures of macrophytes (Lovvorn, 1989; Mitchell et aI., 1994). However, at this point in the
growing season of macrophytes, most aboveground tissue is senescent or soon to
be senescent. Consumption of below ground parts (including nutrient storage and
overwintering structures) might have a greater long-term impact on the macrophytes, but few studies have measured consumption of below ground parts (Korshgen et aI., 1988; Lovvorn, 1989; Michot and Chadwick, 1994), and few, if any,
have looked at the impact on the following year's growth. It is critical to assess the
importance of this common manifestation of herbivory by aquatic birds.
Superimposed on the rebounding populations of many waterfowl, herbivorous
mute swans (Cygnus alar) have been introduced to North America. Mute swans
have spread throughout eastern North America in recent decades, where they may
reduce submerged vegetation through direct consumption, wasteful feeding, and
nest construction. A single territorial breeding pair can uproot up to 0.2 ha of
emergent vegetation for nest construction, although the area affected is typically
smaller (Willey and Halla, 1972).
Thus, the temporal dynamics of aquatic bird populations-from seasonal migrations to decades-long popUlation trends-make it difficult to generalize about
impact on macrophytes in the past or the present. Even infrequent, but intense,
bouts of macrophyte feeding by aquatic birds on a lake might restart succession of
macrophytes, causing a long-term impact the cause of which would be easy to
miss. Enough examples exist, however, of large reductions of macrophytes by
aquatic birds that limnologists can no longer ignore them as potential determinants
of macrophyte abundance and species composition. Much work remains to determine whether the impact of waterfowl is usually small or whether limnologists
must often consider waterfowl in understanding dynamics of macrophytes.
Diet Specialization and a Model of Plant S(~lection
The past bias of limnologists in focusing herbivory studies on insects rather than
other herbivores has been misleading for at least two reasons. First, insects usually
cause much less damage than other herbivorous taxa (Fig. 8.2). Second, most
terrestrial (Strong et aI., 1984; Bernays, 1989) and freshwater (Newman, 1991)
insects found on vascular plants are oligophagous. Although freshwater macrophyte-eating insects are less oligophagous on average than terrestrial insects
162
D.M. Lodge et al.
TabIeS.t. Degree of Specialization Among Freshwater Taxa That Are Primarily HeIbivorous
on Macrophytesa
Plants eaten per herbivore species
Insectsb
MammalsC
Birdsd
Turtlese
Fishes!
Crayfishesg
Plants as % of diet
Families
Genera
?-100%
NA
43-99%
27-89%
20-95%
12-80%
1-3
4-7
7-19
NA
9
>11
<3
NA
9-32
4-8
NA
NA
aFor birds, data are for taxa in Figure 8.3. NA, not available.
hoata from Newman, 1991.
CData from Henderson, 1960; Whitaker, 1980; Jones and Birney, 1988; Lacki et aI., 1990; Wilsey
and Chabreck, 1991; Shaffer et aI., 1992; Doucet and Fryxell, 1993.
d Data from Fasset, 1957; Willey and Halla, 1972; Palmer, 1976; Mitchell, 1994.
eData from Carr, 1952; Conant, 1958; Clark and Gibbons, 1969; Minton, 1972; Mount, 1975; Behler
and King, 1979; Parmenter, 1980; Ernst, 1983; Parmenter and Avery, 1990.
fData from Nichols, 1991; Bain, 1993.
8Data from Lodge and Hill, 1994; Lodge and Cronin, unpublished data.
(Newman, 1991), most aquatic insects are still far more specialized than are other
aquatic herbivore. taxa (Table 8.1). Because (1) the herbivore taxa that cause the
most damage (Fig. 8.2) are generalists (Table 8.1), (2) generalist herbivores often
change the relative abundance of macrophyte species (Lodge, 1991), and (3)
macrophyte community composition, in addition to macrophyte abundance, affects the ecosystem role of macrophytes (Carpenter and Lodge, 1986; van Donk,
this volume, Chapter 19), it becomes important to understand which plants will be
most affected by herbivores and how macrophyte communities will change under
the influence of herbivory.
Model of Plant Selection by Generalist Herbivores
Although few direct tests of the basis of plant selection by freshwater herbivores have
been conducted, enough evidence has accumulated to suggest a conceptual model to
guide further work (Fig. 8.4). To be preferentially consumed, a macrophyte must
(1) have a structure (morphology, toughness, and surface features) that makes it
possible for an herbivore to take a bite; (2) lack chemical deterrents; and (3) be
nutritious. Evidence for each element of this model is described briefly below.
Plant Structure
For crayfishes, the freshwater herbivore for which the most experiments addressing plant selection have been conducted, emergent plants are typically much less
reduced than submerged plants (Chambers et al., 1991; Lodge, 1991; Cronin, this
8. Impact of Herbivory on Plant Standing Crop
163
No
not eaten
little eaten
eat if nothing else available
+Yes
highly preferred
Figure 8.4. Conceptual model of how diet composition is determined in freshwater
herbivores.
volume, Chapter 21). This is also true for Limnephilus caddis fly larvae (Lodge,
unpublished data). For both taxa, low consumption of emergent plants seems to result
from the structure of the plants (i.e., toughness and/or the mismatch between small
mouthparts and broad, flat plant surfaces). When the same plants are freeze-dried,
ground, and reconstituted in an alginate gel, they often become highly preferred,
showing clearly the deterrent quality of plant structure (Cronin, this volume, Chapter
21).
Trichomes, thick cuticles, and other surface structures that defend many terrestrial plants from herbivores are largely absent from submerged aquatic plants
(Levin, 1973; Grubb, 1986). This probably results from selection pressure to
minimize boundary layers to increase gas diffusion across the submerged plant
surface. Thus, submerged macrophytes may require alternative deterrents.
Chemical Deterrents and Stimulants
Many authors have suggested that chemical deterrents and attractants playa role
in the macrophyte-herbivore interaction (Otto, 1983; Sterry et aI., 1983; Haynes
and Taylor, 1984; Suren and Lake, 1989; Jefferies, 1990; Center and Wright, 1991;
Lodge, 1991; Newman, 1991). Nevertheless, only one example exists of an identified chemical defense (glucosinolate-myrosinase system) that deters herbivores
from eating an aquatic plant, watercress (Nasturtium officinale) (Newman et aI.,
1992). However, in many freshwater macrophytes, many classes of compounds
exist that are known to be deterrent to many terrestrial herbivores (Lodge, 1991;
Rosenthal and Berenbaum, 1992). More important, examples are mounting of
unidentified plant compounds that deter freshwater herbivores (Buchsbaum et al.,
1984; Cronin, this volume, Chapter 21).
Nutritional Value
Herbivores are, in general, nitrogen limited, and thus many terrestrial herbivores
preferentially consume high-nitrogen plants (Mattson, 1980). Although supporting
164
D.M. Lodge et al.
data for freshwater habitats are still few and primarily correlative, the same pattern
appears to be true for freshwater herbivores. A snail, three species of caddisflies,
and an amphipod prefer high-nitrogen green watercress tissue over low-nitrogen
yellowed watercress, as long as the chemical defense mechanism in the green
tissue is inoperative (Newman et aI., 1996). The crayfish Procambarus also
prefers high-nitrogen plants among undefended species (Cronin, this volume,
Chapter 21). Species selection by the crayfish Orconectes can be reversed by
reversing the nitrogen content of different submerged macrophyte species (Lodge
et aI., unpublished data). These examples are all consistent with the model suggesting that among plants that are neither structurally nor chemically defended,
nitrogen content may often determine plant selection (Fig. 8.4).
Implications for Herbivore Impact on Macrophyte Species Composition
The specific predictions of this model will differ to some extent among herbivore
taxa because, for example, a plant that is structurally defended against caddisflies
may be handled easily by swans. In addition, other factors important to herbivores
may differ among plants and affect herbivore impact (e.g., predation refuge
offered by the plant for the herbivore [Duffy and Hay, 1994], satiation of the
herbivore [Cronin and Hay, 1996a], and feeding history of the herbivore [provenza, 1995]). Nevertheless, the model (Fig. 8.4) is consistent with the data known to
us and provides a useful framework for additional experimental and observational
work on the impact of herbivores on macrophyte communities.
Conclusions and Role of Herbivores in Stabilizing-Destabilizing
Alternate Lake States
From our summary of the literature on herbivory, it is clear that herbivores often
reduce plant standing stock in freshwater habitats, as they do in terrestrial and
marine habitats. To the limited extent that limnologists have studied herbivory on
macrophytes, they have given undue attention to insects and insufficient attention
to the role of larger herbivores-especially birds, fishes, and crayfishes-in determining macrophyte abundance in lakes. Increasing abundance of aquatic birds and
mammals in both North America and Europe makes them especially deserving of
studies of their long-term impact on macrophyte abundance. In addition, theory
predicts that the generalist nature of plant selection characteristic of the understudied herbivores makes them potentially more powerful suppressers of overall
macrophyte abundance than more specialized insect herbivores (Murdoch and
Bence, 1987).
Any herbivore that substantially reduces macrophyte standing stock in shallow
lakes could playa role in destabilizing any clearwater macrophyte-dominated state
(Scheffer et al., 1993). Ample evidence already exists of the importance of European coots in suppressing the recovery of macrophytes (Lake Stigsholm; S!/Indergaard et al., 1996) or reducing macrophyte abundance (Lake Zwemlust; van Donk
8. Impact of Herbivory on Plant Standing Crop
165
and Otte, 1996; van Donk, this volume, Chapter 19) and thereby resisting the
establishment of the macrophyte-dominated state or the reversal of the phytoplankton-dominated state. In addition, coots were apparently responsible for a
shift in macrophyte community composition from Elodea (the evergreen nature of
which enhanced the clearwater state) to Potamogeton (the seasonal nature of
which was less effective in maintaining the clearwater state) (van Donk, this
volume, Chapter 19). Thus, if large generalist herbivores such as aquatic birds
continue to increase, they may become an impediment to managing lakes for the
clearwater state.
An additional reason that herbivory may be more important now than in the
past in reducing macrophytes and destabilizing the clearwater state is that
eutrophication has both decreased the occurrence of nutrient conditions under
which the clearwater state can prevail and increased the abundance of fishes that
help maintain the turbid water state in Europe (Persson et al., 1991) and North
America (Laird and Page, 1996). If the boundary between alternative lake states
approximates a threshold (Scheffer and Jeppesen, this volume, Chapter 31), a
small increase in herbivory by fishes or birds (or any other herbivore) could tip the
balance toward the turbid phytoplankton-dominated state. Thus, more rigorous
experiments testing the impact of different herbivores and more observational
work on the abundance, feeding rates, and plant selection of a variety of freshwater herbivores will produce insight on the plant-herbivore interaction that has
immediate implications for lake management.
Acknowledgments. We thank Laura Eidietis for assistance with the literature
review and analysis and for preparation of the bibliography. NSFDEB 94-08452
(D.M.L.) has supported the recent research on herbivory by D.M.L., G.C, and A.F.
A.F. has benefited from a NSF Graduate Research Traine{:ship (NSFGER 945265-001). Dee Butler (USFWS) gave A.F. much assistance in locating waterfowl
data sources. G.c. and D.M.L. thank Mark Hay for many stimulating discussions
about plant-animal interactions. For many helpful suggestions on the manuscript,
we thank Mark Hoyer, Robert McIntosh, and Stuart Mitchell.
References
Abrahamsson, S.AA Dynamics of an isolated population of the crayfish Astacus astacus
Linne. Oikos 17:96-107; 1966.
Allin, C.c.; Chasko, G.c.; Husband, T.P. Mute swans in the Atlantic flyway: a review of the
history, population growth and management needs. Trans. N. E. Sec. Wildl. Soc. 44:3247; 1987.
Anderson, M.G.; Low, J.P. Use of sago pondweed by waterfowl on the Delta Marsh,
Manitoba. J. Wild1. Manage. 40:233-242; 1976.
Andrew, N.L.; Underwood, AJ. Density-dependent foraging in the sea urchin Centrostephanus rodgersii on shallow subtidal reefs in New South Wales, Australia. Mar. Bco1.
Prog. Ser. 99:89-98; 1993.
Bain, M.B. Assessing impacts of introduced aquatic species: grass carp in large systems.
Environ. Manage. 17:211-224; 1993.
Baldassare, G.A; Bolen, E.G. Waterfowl ecology and management. New York: John Wiley
& Sons; 1994.
166
D.M. Lodge et a1.
Banko, W.E. The trumpeter swan: its history, habits and population in the United States.
North American Fauna 63. Washington, DC: Bur. Sport Fish and Wild1.; 1960.
Bazely, D.R.; Jeffries, R.L. Changes in the composition and standing crop of salt-marsh
communities in response to the removal of a grazer. J. Eco1. 74:693-706; 1986.
Behler, J.L.; King, EW. The Audubon Society field guide to North American reptiles and
amphibians. New York: Knopf; 1979.
Bellrose, Ee. Ducks, geese & swans of North America. Harrisburg, PA: Stackpole Books;
1980.
Benedetti-Cecchi, L.; Cinelli, E Early patterns of algal succession in a midlittoral community of the Mediterranean sea: a multifactorial experiment. J. Exp. Mar. BioI. Ecol.
169:15-31; 1993.
Bernays, E.A. Host range in phytophagous insects: the potential role of generalist predators.
Evo1. Eco1. 3:299-311; 1989.
Bertness, M.D. Habitat and community modification by an introduced herbivorous snail.
Ecology. 65:370--381; 1984.
Bickart, K.J. Recent advances in the study of neogene fossil birds I. The birds of the late
Miocene-early Pliocene Big Sandy Formation, Mohave County, Arizona. Ornitho1.
Monogr. 44:1-72; 1990.
Bonser, S.P.; Reader, RJ. Plant competition and herbivory in relation to vegetation biomass.
Ecology 76:2176-2183; 1995.
Bronmark, C. Effects of tench and perch on interactions in a freshwater, benthic food chain.
Ecology 75:1818-1828; 1994.
Bronmark, C. How do herbivorous freshwater snails affect macrophytes?-a comment.
Ecology 71:1212-1215; 1990.
Buchsbaum, R; Valiela, I.; Swain, T. The role of phenolic compounds and other plant
constitutents in feeding by Canada geese in a coastal marsh. Oecologia 63:343-349;
1984.
Buschmann, AH. The role of herbivory and desiccation on early successional patterns of
intertidal macroalgae in southern Chile. J. Exp. Mar. BioI. Eco1. 139:221-230; 1990.
Caithamer, D.E; Dubovsky, J.A Waterfowl popUlation status, 1995. Laurel, MD: U.S. Fish
and Wild1. Ser.; 1995.
Campbell, H.W.; Irvine, AB. Feeding ecology of the West Indian manatee Trichechus
manatusLinnaeus. Aquaculture 12:249-251; 1977.
Cargill, S.M.; Jefferies, RL. The effects of grazing by lesser snow geese on the vegetation
of a sub-arctic salt marsh. J. Aquat. Eco1. 21 :669-686; 1984.
Cargill, S.M.; Jefferies, R.L. Changes in the composition and standing crop of salt-marsh
communities in response to the removal of a grazer. J. Ecol. 74:693-706; 1986.
Carpenter, R.C. Grazing by Diadema antillarum (Philippi) and its effects on the benthic
algal community. J. Mar. Res. 39:749-765; 1981.
Carpenter, R.e. Partitioning herbivory and its effects on coral reef algal communities. Ecol.
Monogr. 56:345-363; 1986.
Carpenter, S.R; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Carpenter, S.R.; Titus, J.E. Composition and spatial heterogeneity of submersed vegetation
in a softwater lake in Wisconsin. Vegetatio 57:153-165; 1984.
Carr, A Handbook of turtles. Ithaca, NY: Comstock Publishing; 1952.
Carter, Y.; Rybicki, N.B. The effects of grazers and light penetration on the survival of
transplants of Vallisneria americana Michx. in the tidal Potomac River, Maryland.
Aquat. Bot. 23:197-213; 1985.
Castenholz, RW. The effect of grazing on marine littoral diatom populations. Ecology
42:783-794; 1961.
Cattaneo, A; Mousseau, B. Empirical analysis of the removal rate of periphyton by grazers.
Oecologia 103:249-254; 1995.
8. Impact of Herbivory on Plant Standing Crop
167
Center, T.D.; Wright, A.D. Age and phytochemical composition of waterhyacinth (Pontederiaceae) leaves determine their acceptability to Neochetina eichhomiae (Coleoptera:
Curculionidae). Environ. Entomol. 20(1):323-334; 1991.
Chambers, P.A.; Hanson, J.M.; Prepas, E.E. The effect of aquatic plant chemistry and
morphology on feeding selectivity by the crayfish, Orconectes virilis. Freshwat. BioI.
25:339-348; 1991.
Chikwenhere, G.P. Biological control of water lettuce in various impoundments in Zimbabwe. J. Aquat. Plant Manage. 32:27-29; 1994.
Clark, D.B.; Gibbons, J.W. Dietary shift in the turtle Pseudemys scripta (Schoepf±) from
youth to maturity. Copeia 4:704-706; 1969.
Coley, P.D.; Aide, T.M. Comparison of herbivory and plant defenses in temparate and
tropical broad-leaved forests. In: Price, P.W.; Lewinsohn, T.M.; Fernandes, G.W.; Benson, W.W., eds. Plant-animal interactions: evolutionary ecology in tropical and temparate regions. New York: John Wiley & Sons; 1991:25-49.
Conant, R. A field guide to reptiles and amphibians. Boston: Houghton Mifflin; 1958.
Conover, M.R.; Kania, G.S. Impact of interspecific aggression and herbivory by mute
swans on native waterfowl and aquatic vegetation in New England. Auk 3:744-748;
1994.
Coppock, D.L.; Detling, J.K.; Ellis, J.E.; Dyer, M.1. Plant-herbivore interactions in a North
American mixed-grass prairie. I. Effects of black-tailed prairie dogs on intraseasonal
aboveground plant biomass and nutrient dynamics and plant species diversity. Oecologia 56: 1-9; 1983.
Creed, R.P. Direct and indirect effects of crayfish grazing in a streanl community. Ecology
75:2091-2103; 1994.
Cronin, G.; Hay, M. Seaweed-herbivore interactions depend on recent history of both the
plant and animal. Ecology 77:1531-1543; 1996a.
Cronin, G.; Hay, M. Within-plant variation in seaweed palatability and chemical defenses:
optimal defense theory versus the growth-differentiation balance hypothesis. Oecologia
105:361-368; 1996b.
Crutchfield, J.U.; Schiller, D.H.; Herlong, D.D.; Mallin, M.A. Establishment and impact of
redbelly tilapia in a vegetated cooling resevoir. J. Aquat. Plant. Manage. 30:28-35;
1992.
Cyr, H.; Pace, M.L. Magnitude and patterns of herbivory in aquatic and terresrial ecosystems. Nature 361:148-150; 1993.
Dahl, T.E. Wetlands losses in the United States, 1780's to 1980's. Washington, DC: U.S.
Fish and Wildl. Serv.; 1990.
Dean, J.L. Biology of the crayfish Orconectes causeyi and its control of aquatic weeds in
Trout Lake. Tech. Paper, Bureau Sport Fish Wildl. 24:1-15; 1969.
Dirzo, R.; Miranda, A. Altered patterns of herbivory and diversity in the forest understory:
a case study of the possible consequences of contemporary defaunation. In: Price, P. W.;
Lewinsohn, T.M.; Fernandes, G.W.; Benson, W.W., eds. Plant-animal interactions:
evolutionary ecology in tropical and temparate regions. New York: John Wiley & Sons;
1991 :273-287.
Doucet, C.M.; Fryxell, J.M. The effect of nutritional quality on forage preference by
beavers. Oikos 67:201-208; 1993.
Duarte, C.M. Nutrient concentration of aquatic plants: patterns across species. Limnol.
Oceanogr. 37:882-889; 1992.
Duffy, J.E.; Hay, M.E. Herbivore resistance to seaweed chemical defense: the roles of
mobility and predator risk. Ecology 75:1304-1319; 1994.
Duggins, D.O. Kelp beds and sea otters: an experimental approach. Ecology 61:447-453;
1980.
Ernst, C. Turtles of the United States. Washington, DC: Smithsonian Institute Press; 1972.
Esler, D. An assessment of American coot herbivory of hydrilla. J. Wildl. Manage. 53:
1147-1149; 1989.
168
D.M. Lodge et al.
Estes, I.A.; Duggins, D.O. Sea otters and kelp forests in Alaska: generality and variation in
a community ecological paradigm. Eco!. Monogr. 65:75-100; 1995.
Estes, J.A.; Steinberg, P.D. Predation, herbivory and kelp evolution. Paleobiology 14: 1936; 1988.
Fassett, N.C. A manual of aquatic plants. Madison, WI: University of Wisconsin Press;
1957.
FeminelJa, I.W.; Hawkins, c.A. Interactions between stream herbivores and periphyton: a
quantitative analysis of past experiments. J. North Am. Benth. Soc. 14:465-509; 1995.
Feminella, J.W.; Resh, V.H. Submersed macrophytes and grazing crayfish: an experimental
study of herbivory in a California freshwater marsh. Holarct. Eco!. 12: 1-8; 1989.
Fletcher, W.1. Interactions among subtidal Australian sea urchins, gastropods, and algae:
effects of experimental removal. Ecol. Monogr. 57:89-109; 1987.
Foster, S.A. The relative impacts of grazing by Caribbean coral reef fishes and Diadema:
effects of habitat and surge. 1. Exp. Mar. BioI. Ecol. 15: 1-20; 1987.
Gaines, S.D.; Lubchenco, J. A uniiied approach to marine plant-herbivore interactions. II.
Biogeography. Annu. Rev. Eco!. Syst. 13: 111-138; 1982.
Geller, J.B. Gastropod grazers and algal colonization on a rocky shore in northern California: the importance of body size of grazers. J. Exp. Mar. BioI. Eco!. 150:1-7; 1991.
Gillette, L.N.; Shea, R. An evaluation of trumpeter swan management today and a vision for
the future. Trans. 60th North Am. Wildl. & Natur. Resour. Conf. Washington, DC:
Wildlife Management Institute; 1995:258-265.
Giroux, J.F.; Bedard, J. The effects of grazing by greater snow geese on the vegetation of
tidal marshes in the St. Lawrence Estuary. J. Appl. Ecol. 24:773-788; 1987.
Grubb, P.J. Sc1erophylls, pachyphylls, and pycnophylls: the nature and significance of hard
leaf surfaces. In: Juniper, B.E.; Southwood, T.R.E .. eds. Insects and the plant surface.
London: Edwin Arnold; 1986:137-150.
Hansson, L.; Johansson, L.; Persson, L. Effects of fish grazing on nutrient release and
succession of primary producers. Limnol. Oceanogr. 32:723-729; 1987.
Hart, D.D. Community organization in streams: the importance of species interactions,
physical factors, and chance. Oecologia 91 :220--228; 1992.
Hatcher, B.G. The interaction between grazing organisms and the epilithic algal community
of a coral reef: a quantitative assessment. Proc. 4th lnt. Coral Reef Congr. 2; 1981 :515524.
Hatcher, B.G.; Larkum, A.W.D. An experimental analysis of factors controlling the standing crop of the epilithic algal community on a coral reef. J. Exp. Mar. BioI. Ecol.
69:61-84; 1983.
Hay, M.E. Herbivory, algal distribution, and the maintenance of between-habitat diversity
on a tropical fringing reef. Am. Nat. 118:520-540; 1981.
Hay, M.E. Predictable spatial escapes from herbivory: how do these affect the evolution of
herbivore resistance in tropical marine communities? Oecologia 64:396-407; 1984.
Hay, M.E. Fish-seaweed interactions on coral reefs: effects of herbivorous fishes and
adaptations of their prey. In: Sale, P.F., ed. The ecology of fishes on coral reefs. San
Diego, CA: Academic Press; 1991:96-119.
Hay, M.E.; Goertemiller, T. Between-habitat differences in herbivore impact on Caribbean
coral reefs. In: Reake, M.L., ed. The ecology of deep and shallow coral reefs. Symposia
Series for Undersea Research. Vol. I. Rockville, MD: Office of Undersea Research,
NOAA; 1983:97-102.
Hay, M.E.; Taylor, P.R. Competition between herbivorous fishes and urchins on Caribbean
reefs. Oecologia 65:591-598; 1985.
Hay, M.E.; Colburn, T.; Downing, D. Spatial and temporal patterns in herbivory on a
Caribbean fringing reef: the effects on plant distribution. Oecologia 58:299-308; 1983.
Haynes, A.; Taylor, B.1.R. Food finding and food preference in Potamopyrgus jenkinsi
(E.A. Smith) (Gastropoda: Prosobranchia). Arch. Hydrobiol. 100:479-491; 1984.
8. Impact of Herbivory on Plant Standing Crop
169
Heck, K.L.; Valentine, J.E Sea urchin herbivory: evidence for long-lasting effects in
subtropical seagrass meadows. J. Exp. Mar. BioI. Ecol. 189:205-217; 1995.
Henderson, R.E Beaver in Kansas. Lawrence, KS: State Biological Survey and Museum of
Natural History; 1960.
Hengeveld, R. Dynamics of biological invasions. New York: Chapman & Hall; 1989.
Herms, D.A; Mattson, W.J. The dilemma of plants: to grow or defend. Q. Rev. BioI.
67:283-335; 1992.
Himmelman, J.H.; Cardinal, A; Bourget, E. Community development following removal
of urchins, Strongylocentrotus droebachiensis, from the rocky subtidal zone of the St.
Lawrence Estuary, Eastern Canada. Oecologia 59:27-39; 1983.
Hoyer, M.V.; Canfield, D.E. Bird abundance and species richness on Florida lakes: influence of trophic status, lake morphology, and aquatic macrophytes. Hydrobiologia
2791280:107-119; 1994.
Hughes, T.P.; Reed, D.C.; Boyle, M.J. Herbivory on coral reefs: community structure
following mass mortalities of sea urchins. J. Exp. Mar. BioI. Ecol. 113:39-59; 1987.
Hulme, P.E. Herbivores and the performance of grassland plants: a comparison of arthropod, mollusc and rodent herbivory. J. Ecol. 84:43-51; 1996.
Huntley, N.J. Influence of refuging consumers (Pikas: Ochotona princeps) on subalpine
meadow vegetation. Ecology 68:274-283; 1987.
Jacobsen, D.; Sand-Jensen, K. Herbivory of invertebrates on submerged macrophytes from
Danish freshwaters. Freshwat. BioI. 28:301-308; 1992.
Jacobsen, D.; Sand-Jensen, K. Invertebrate herbivory on the submerged macrophyte Potamogeton perfoliatus in a Danish stream. Freshwat. BioI. 31 :43-52; 1994.
Jacobsen, D.; Sand-Jensen, K. Variability of invertebrate herbivory on the submerged
macrophyte Potamogeton perfoliatus. Freshwat. BioI. 34:357-365; 1995.
Jefferies, M. Evidence of induced plant defences in a pondweed. Freshwat. BioI. 23:265269; 1990.
Jones, G.P. Interactions between herbivorous fishes and macroalgae on a temperate rocky
reef. J. Exp. Mar. BioI. Ecol. 159:217-235; 1992.
Jones, J.K.; Birney, E.e. Handbook of mammals of the north central states. Minneapolis,
MN: University of Minnesota Press; 1988.
Julien, M.H.; Bourne, AS.; Low, V.H.K. Growth of the weed Alternanthera philoxeroides
(Martius) Grisebach, (alligator weed) in aquatic and terrestrial habitats. Aust. Plant
Protection Q. 7:102-108; 1992.
Jupp, B.P.; Spence, D.H.N. Limitations of macrophytes in an eutrophic lake, Loch Leven. J.
Ecol. 65:431-446; 1977.
Kantrud, H.A; Stewart, R.E. Use of natural basin wetlands by breeding waterfowl in North
Dakota. J. Wildl. Manage. 41:243-253; 1977.
Keats, D.W.; South, G.R.; Steele, D.H. Effects of an experimental reduction in grazing by
green sea urchins on a macroalgal community in eastern Newfoundland. Mar. Ecol.
Prog. Ser. 68:181-193; 1990.
Kiorboe, T. Distribution and production of submerged macrophytes in Tipper Grund (Ringkobing Fjord, Denmark), and the impact of waterfowl grazing. J. Appl. Ecol. 17:675687; 1980.
Kornij6w, R. The importance of invertebrates as consumers of freshwater macrophytes.
Wiad. Ekol. 40:181-195; 1994.
Kornij6w, R. Cumulative consumption of the lake macrophyte Elodea by abundant generalist invertebrate herbivores. Hydrobiologia 319: 185-190; 1996.
Korshgen, e.E.; George, L.S.; Green, W.L. Feeding ecology of canvasbacks staging on pool
7 of the upper Mississippi River in waterfowl in winter. In: Weller, M.W., ed. Minneapolis, MN: University of Minnesota Press; 1988:237-249.
Lacki, M.J.; Peneston, W.T.; Adams, K.B.; Vogt, ED.; Houppert, J.e. Summer foraging
patterns and diet selection of muskrats inhabiting a fen wetland. Can. J. Zool. 68: 11631167; 1990.
170
D.M. Lodge et al.
Laird, e.A; Page, L.M. Non-native fishes inhabiting the streams and lakes of Illinois. Ill.
Nat. History Surv. Bull. 35:1-51; 1996.
Lamberti, G.A The role of periphyton in benthic food webs. In: Stevenson, R.I.; Bothwell,
ML; Lowe, RL, eds. Algal ecology: freshwater benthic ecosystems. New York:
Academic Press; 1996:533-573.
Levin, D.A The role oftrichomes in plant defense. Q. Rev. BioI. 48:3-15; 1973.
Lewis, S.M. The role of herbivorous fishes in the organization of a Caribbean reef community. Ecol. Monogr. 56:183-200; 1986.
Lewis, S.M.; Norris, J.N.; Searles, R.B. The regulation of morphological plasticity in
tropical reef algae by herbivory. Ecology 68:636--641; 1987.
Lodge, D.M. Herbivory on freshwater macrophytes. Aquat. Bot. 41:195-224; 1991.
Lodge, D.M.; Hill, AM. Factors governing species composition, population size, and
productivity of cool-water crayfishes. Nordic J. Freshwat. Res. 69:111-136; 1994.
Lodge, D.M.; Lorman, J.G. Reductions in submersed macrophyte biomass and species
richness by the crayfish Orconectes rusticus. Can. J. Fish. Aquat. Sci. 44:591-597;
1987.
Lodge, D.M.; Krabbenhoft, D.P.; Striegl, R.G. A positive relationship between groundwater
velocity and submersed macrophyte biomass in Sparkling Lake, Wisconsin. Lirnnol.
Oceanogr. 34:235-239; 1989.
Lodge, D.M.; Kershner, M.W.; Aloi, J.E.; Covich, AP. Effects of an omnivorous crayfish
(Orconectes ructicus) on a freshwater littoral food web. Ecology 75: 1265-1281; 1994.
Lodge, D.M.; Stein, R.A; Brown, K.M.; Covich, AP.; Bronmark, e.; Garvey, J.E; Klosiewski, S.P. Predicting impact of freshwater exotic species on native biodiversity:
challenges in spatial scaling. Aust. J. Ecol. (in press).
Lovvorn, J.R. Distribution responses of canvasback ducks to weather and habitat change. J.
Appl. Ecol. 26:113-130; 1989.
Lubchenco, J.; Gaines, S.D. A unified approach to marine plant-herbivore interactions. I.
Populations and communities. Annu. Rev. Ecol. Syst. 12:405-437; 1981.
Maceina, M.J.; Cichra, M.E; Betsill, R.K.; Bettoli, P.w. Lirnnological changes in a large
reservoir following vegetation removal by grass carp. J. Freshwat. Eco1. 7:81-95; 1992.
Maringer, J.; Bandi, H.G. Art in the ice age. New York: Frederick A Praeger; 1953.
Marquis, R.J.; Whelan, e.J. Insectivorous birds increase growth of white oak through
consumption of leaf-chewing insects. Ecology 75:2007-2014; 1994.
Martin, T.H.; Crowder, L.B.; Dumas, C.F.; Burkholder, J.M. Indirect effects of fish on
macrophytes in Bays Mountain Lake: evidence for a littoral trophic cascade. Oecologia
89:476-481; 1992.
Matthews, M.; Reynolds, J.D. Ecological impact of crayfish plague in Ireland. Hydrobiologia 234: 1-6; 1992.
Matthews, W.J.; Stewart, AJ.; Power, M.E. Grazing fishes as components of North American stream ecosystems: Effects of Campostoma anomalum. In: Matthews, W.I.; Heins,
D.C., eds. Community and evolutionary ecology of North American stream fishes.
Norman, OK: University of Oklahoma Press; 1987: 128-135.
Mattson, W.J., Jr. Herbivory in relation to plant nitrogen content. Annu. Rev. Eco1. Syst.
11:119-161; 1980.
McAtee, w.L. Three important wild duck foods. U.S. Bur. BioI. Surv. Circ. 81; 1911.
McNaughton, S.J. Ecology of a grazing ecosystem: the Serengeti. Ecol. Monogr. 55:259294; 1985.
Michot, T.e.; Chadwick, P.C. Winter biomass and nutrient values of three seagrass species
as potential foods for redheads (Aythya americana Eyton) in Chandeleur Sound, Louisiana. Wetlands 14:276-283; 1994.
Miller, M.W.; Hay, M.E. Coral-seaweed-grazer-nutrient interactions on temperate reefs.
Eco1. Monogr. 66:323-344; 1996.
Minton, S.A Amphibians and reptiles of Indiana. Indianapolis, IN: Indiana Academy of
Science; 1972.
8. Impact of Herbivory on Plant Standing Crop
171
Mitchell, e.A.; Thomas, W.e.; Zwank, P.J. Herbivory on shoalgrass by wintering redheads
in Texas. J. Wildl. Manage. 58:131-141; 1994.
Mitchell, CD. Trumpeter swan (Cygnus buccinator). In: Poole, A.; Gill, P., eds. The birds
of North America, No. 105. Philadelphia, PA: The Academy of Natural Sciences;
Washington, DC: The American Ornithologists' Union; 1994.
Mitchell, S.P.; Wass, R.T. Quantifying herbivory: grazing consumption and interaction
strength. Oikos 77:1-4; 1996a.
Mitchell, S.P.; Wass, R.T. Grazing by black swans (Cygnus atratus Lathram), physical
factors, and the growth and loss of aquatic vegetation in a shallow lake. Aquat. Bot.
55:205-215; 1996b.
Mitchell, S.P.; Hamilton, D.P.; Macgibbon, W.S.; Nayer, P.K.B.; Reynolds, R.N. Interrelations between phytoplankton, submerged macrophytes, black swans (Cygnus atratus)
and zooplankton in a shallow New Zealand lake. Int. Rev. Ges. Hydrobiol. 73: 145-170;
1988.
Morrison, D. Comparing fish and urchin grazing in shallow and deeper coral reef algal
communities. Ecology 69:1367-1382; 1988.
Mount, R.H. Reptiles and amphibians of Alabama. Auburn, AL: Auburn Printing Company;
1975.
Murdoch, W.W.; Bence, J. General predators and unstable prey populations. In: Kerfoot,
W.e.; Sih, A., eds. Predation: direct and indirect impacts on aquatic communities.
Hanover, NH: University Press of New England; 1987:17-30.
Newman, R.M. Herbivory and detritivory on freshwater macrophytes by invertebrates: a
review. J. North Am. Benth. Soc. 10:89-114; 1991.
Newman, R.M.; Hanscom, Z.; Kerfoot, W.C. The watercress glucosinolate-myrosinase
system: a feeding deterrent to caddisflies, snails and amphipods. Oecologia 92: 1-7;
1992.
Newman, R.M.; Kerfoot, W.C; Hanscom, Z.A.e. Watercress allelochemical defends high
nitrogen foliage against consumption: effects on freshwater invertebrate herbivores.
Ecology 77:2312-2323; 1996.
Nichols, S.A. The interaction between biology and the management of aquatic macrophytes. Aquat. Bot. 41:225-252; 1991.
Ogden, J.C; Brown, R.A.; Salesky, N. Grazing by the echinoid Diadema antillarum
Philippi: formation of halos around West Indian patch reefs. Science 182:715-717;
1979.
Otto, C. Adaptations to benthic freshwater herbivory. In: Wetzel, R.G., ed. Periphyton of
freshwater ecosystems. Boston: Dr. W. Junk Publ.; 1983:199-205.
Owen-Smith, N.; Cooper, S.M. Palatability of woody plants to browsing ruminants in a
South African savanna. Ecology 68:319-331; 1987.
Painter, D.S.; McCabe, K.J. Investigation into the disappearance of Eurasian water-milfoil
from the Kawartha Lakes. J. Aquat. Plant Manage. 26:3-12; 1988.
Palmer, R.S. Handbook of North American birds. New Haven, CT: Yale University Press;
1976.
Parmenter, R.R. Effects of food availability and water temperature on the feeding ecology
of pond sliders. Copeia 3:503-514; 1980.
Parmenter, R.R.; Avery. 1990. The feeding ecology of the slider turtle. In: Parmenter, R.R.;
Avery, H.W., eds. Life history and ecology of the slider turtle. Washington, DC: Smithsonian Institution Press; 1990:257-265.
Pelikan, J.; Svoboda, J.; Kvet, J. Relationship between the populations of muskrats (Ondatra zibethica) and the primary production of cattail (Typha latifolia). Hydrobiologia
12:177-180; 1971.
Persson, L.; Diehl, S.; Johansson, L.; Andersson, G.; Harnrin, S.P. Shifts in fish communities along the productivity gradient of temperate lakes-patterns and the importance of size-structured interactions. J. Fish. BioI. 38:281-293; 1991.
172
D.M. Lodge et aI.
Polunin, N. V.C.; Klumpp, D.W. Algal food supply and grazer demand in a very productive
coral reef zone. J. Exp. Mar. BioI. EcoI. 164:1-15; 1992.
Power, M.E.; Stewart, AJ.; Matthews, WJ. Grazer control of algae in an Ozard Mountain
stream: effects of short-term exclusion. Ecology 69: 1894-1898; 1988.
Prejs, A Herbivory by temperate freshwater fishes and its consequences. Environ. BioI.
Fish. 10:281-296; 1984.
Prince, J. Limited effects of the sea urchin Echinometra mathaei (de Blainville) on the
recruitment of benthic algae and macroinvertebrates into intertidal rock platforms at
Rottnest Island, Western Australia. l Exp. Mar. BioI. EcoI. 186:237-258; 1995.
Provenza, F.D. Tracking variable environments: there is more than one kind of memory.
J. Chem. Ecol. 21:911-924; 1995.
Randall, J.E. Overgrazing of algae by herbivorous marine fishes. Ecology 42:812; 1961.
Randall, J.E. Grazing effect on seagrasses by herbivorous reef fishes in the West Indies.
Ecology 46:255-260; 1965.
Reinecke, KJ.; Kaminski, R.M.; Moorhead, D.J.; Hodges, lD.; Nassar, lR. Mississippi
Alluvial Valley. In: Smith, L.M.; Pederson, R.L.; Kaminski, R.M., eds. Habitat management for migrating and wintering waterfowl in North America. Lubbock, TX: Texas
Tech University Press; 1989.
Rhoades, D.F. Offensive-defensive interactions between insects and plants: their relevance
in herbivore population dynamics and ecological theory. Am. Nat. 125:205-238; 1985.
Richardson, M.J.; Whoriskey, F.G.; Roy, L.H. Turbidity generation and biological impacts
of an exotic fish Carassius auratus, introduced into shallow seasonally anoxic ponds.
l Fish. BioI. 47:576-585; 1995.
Risley, L.S.; Crossley, D.A Herbivore-caused greenfall in the southern Appalachians.
Ecology 69:1118-1127; 1988.
Rockwell, R.; Abraham, K.; Jefferies, R. Tundra under siege. Nat. History 105:20-21; 1996.
Rose, P.M. Midwinter waterfowl counts. In: Pose, P.M., ed. Western Paleartic and SouthWest Asia waterfowl census 1994. Slimbridge, Gloucester, UK: Waterfowl and Wetland
Res. Bur. IWRB; 1994.
Rosenthal, GA.; Berenbaum, M.R., eds. Herbivores: their interactions with secondary plant
metabolites. Vol. II: Evolutionary and ecological processes. New York: Academic Press;
1992.
Russ, G.R. Is the rate of removal by grazers reduced inside territories of tropical damselfishes? J. Exp. Mar. BioI. EcoI. 110:1-17; 1987.
Sammarco, P.W. Diadema and its relationship to coral spat mortality: grazing, competition,
and biological disturbance. J. Exp. Mar. BioI. Eco!. 45:245-272; 1980.
Sammarco, P. W. Effects of grazing by Diadema antillarum Philippi (Echinodermata:
Echinoidea) on algal diversity and community structure. J. Exp. Mar. BioI. Eco!.
65:83-105; 1982.
Sammarco, P.W.; Levinton, J.S.; Ogden, J.e. Grazing and control of coral reef community
structure by Diadema antillarum Philippi (Echinodermata: Echinoidea) : a preliminary
study. J. Mar. Res. 32:47-53; 1974.
Santelices, B.; Martinez, E. Effects of filter-feeders and grazers on algal settlement and
growth in mussel beds. J. Exp. Mar. BioI. EcoI. 118:281-306; 1988.
Samelle, 0.; Kratz, K.W.; Cooper, S.D. Effects of an invertebrate grazer on the spatial
arrangement ofa benthic microhabitat. Oecologia 96:208-218; 1993.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evo!. 8:275-279; 1993.
Schreiner, E.G.; Krueger, K.A; Happe, P.J.; Houston, D.B. Understory patch dynamics and
ungulate herbivory in old-growth forests of Olympic National Park, Washington. Can.
J. For. Res. 26:255-265; 1996.
Schutten, J.; van der Velden, A; Smit, H. Submerged macrophytes in the recently freshened
lake system Volkerak-Zoom (The Netherlands), 1987-1991. Hydrobiologia 275/276:
207-218; 1994.
8. Impact of Herbivory on Plant Standing Crop
173
Scott, M.L.; Haskins, lL. Effects of grazing by chrysomeiid beetles on two wetland
herbaceous species. Bull. Torrey Bot. Club 114:13-17; 1987.
Shaffer, G.P.; Sasser, C.E.; Gosselink, J.G.; Rejmanek, M. Vegetation dynamics in the
emerging Atchatalaya Delta, Louisiana, USA. J. Ecol. 80:677-687; 1992.
Sheldon, S.P. The effects of herbivorous snails on submerged macrophyte communities in
Minnesota lakes. Ecology 68: 1920-1931; 1987.
Sheldon, S.P. More on freshwater snail herbivory: a reply to Brtinmark. Ecology 71: 12151216; 1990.
Shireman, J.V. Control of aquatic weeds with exotic fishes. In: Courtenay, W.R.; Stauffer,
J.R., eds. Distribution, biology, and management of exotic fishes. Baltimore, MD: Johns
Hopkins University Press; 1984: 302-312.
Sih, A. The behavioral response race between predator and prey. Am. Nat. 123:143-150;
1984.
Smith, L.M.; Kadlec, J.A. Fire and herbivory in a Great Salt marsh. Ecology 66:259-265;
1985.
Smith, T.J.; Odum, W.E. The effects of grazing by snow geese on coastal salt marshes.
Ecology 62:98-106; 1981.
S0ndergaard, M.; Bruun, L.; Lauridsen, T.; Jeppesen, E.; Madsen, T.Y. The impact of
grazing waterfowl on submerged macrophytes: in situ experiments in a shallow eutrophic lake. Aquat. Bot. 53:73-84; 1996.
Soska, GJ. Ecological relations between invertebrates and submerged macrophytes in the
lake littoral. Ekol. Pol. 23:393-415; 1975.
Squires, J.R Trumpeter swam food habitats, forage processing, activities and habitat use.
Ph.D. thesis, Univ. Wyoming, Laramie; 1991.
Steinberg, P.D.; Estes, J.A.; Winter, F.e. Evolutionary consequences of food chain length in
kelp forest communities. Proc. Natl. Acad. Sci. U.S.A. 92:8145-8148; 1995.
Steinman, AD. Effects of grazers on freshwater benthic algae. In: Stevenson, RJ.; Bothwell, M.L.; Lowe, R.L., eds. Algal ecology: freshwater benthic eGosystems. New York:
Academic Press; 1996:533-573.
Sterry, P.R; Thomas, J.D.; Patience, R.L. Behavioural responses of Biamphalaria glabrata
(Say) to chemical factors from aquatic macrophytes including decaying Lemna paucicastata (Hegelm ex Engelm). Freshwat. BioI. 13:465-476; 1983.
Strong, D.R.; Lawton, lH.; Southwood, R. Insects on plants: community patterns on plants.
Cambridge, MA: Harvard University Press; 1984.
Suren, A.M.; Lake, P.S. Edibility of fresh and decomposing macrophytes to three species of
freshwater invertebrate herbivores. Hydrobiologia 178: 165-178; 1989.
Taylor, J.N.; Courtenay, W.R.; McCann, I.A. Known impacts of exotic fishes in the
continental United States. In: Courtenay, W.R.; Stauffer, I.R., eds. Distribution, biology,
and management of exotic fishes. Baltimore, MD: Johns Hopkins University Press;
1984:322-373.
Taylor, K.L.; Grace, J .B. The effects of vertebrate herbivory on the plant community structure in
the coastal marshes of the Pearl River, Louisiana, USA. Wetlands 15:68-73; 1995.
Taylor, K.L.; Grace, J.B.; Guntenspergen, G.R.; Foote, A.L. The: interactive effects of
herbivory and fire on an oligohaline marsh, Little Lake, Louisiana, USA. Wetlands
14:82-87; 1994.
Underwood, GJ.C.; Thomas, J.D.; Baker, I.H. An experimental investigation of interactions in snail-macrophyte-epiphyte systems. Oecologia 91 :587-595; 1992.
Urban, E. The mining fauna in four macrophyte species in Mikolajskie Lake. Ekol. Pol.
23:417-438; 1975.
Urbanc, B.O.; Blejec, A. Aquatic macrophytes of Lake Bled: changes in species composition, distribution and production. Hydrobiologia 262: 189-194; 1993.
Valentine, J .F.; Heck, K.L. The role of sea urchin grazing in regulating subtropical seagrass
meadows: evidence from field manipulations in the northern Gulf of Mexico. J. Exp.
Mar. BioI. Ecol. 154:215-230; 1991.
174
D.M. Lodge et al.
Vance, RR Effects of grazing by the sea urchin, Centrostephanus coronatus, on prey
community composition. Ecology 60:537-546; 1979.
van der Velde, G. Initial decomposition of Nymphoides peltata (Gmel.) O. Kuntze (Menyanthaceae), as studied by the leaf-marking method. Hydrobiol. Bull. 16:51-60; 1982.
van Donk, E.; Gulati, R.D. Transition of a lake to turbid state six years after biomanipulation: mechanisms and pathways. Wat. Sci. Techn. 32:197-206; 1995.
van Donk, E.; Otte, A. Effects of grazing by fish and waterfowl on the biomass and species
composition of submerged macrophytes. Hydrobiologia 340:285-290; 1996..
van Donk, E.; De Deckere, E.; Klein-Breteler, J.G.P.; Meulemans, J.T. Herbivory by
waterfowl and fish on macrophytes in a biomanipulated lake: effects on long-term
recovery. Verh. Int. Verein. Limnol. 25:2139-2143; 1994.
van Tamelan, P.G. Early successional mechanisms in the rocky intertidal: the role of direct
and indirect mechanisms. J. Exp. Mar. BioI. Ecol. 112:39-48; 1987.
Verhoeven, J.T.A. Natural regulation of plant biomass in a Ruppia-dorninated system. Proc.
EWRS 5th Symp. on Aquatic Weeds; 1978:53-61.
Vermaat, lE. Periphyton removal by freshwater rnicrograzers. In: van Vierssen, W.; Hootsmans, M.; Vermaat, J.E., eds. Lake Veluwe, a macrophyte-dorninated system under
eutrophication stress. Boston: Kluwer Academic Publishers; 1994:213-249.
Wallace, J.B.; O'Hop, J. Life on a fast pad: waterlilly beetle impact on water lilies. Ecology
66:1534-1544, 1985.
Wanders, J.B.W. The role of benthic algae in the shallow reef of Curacao (Netherlands
Antilles) III: the significance of grazing. Aquat. Bot. 3:357-390; 1977.
Whicker, A.D.; Detling, J.K. Ecological consequences of prairie dog disturbances. Bioscience 38:778-785; 1988.
Whitaker, J.O. The Audubon Society field guide to mammals. New York: Knopf, Inc. 1980.
Willey, C.H.; Halla, B.F. Mute swans of Rhode Island. Wildl. Pamphlet No.8, Rhode Island
Dept. Nat. Res., Div. ofFish and Wildl.; 1972.
Wilsey, B.J.; Chabreck, RH. Nutritional quality of nutria diets in three Louisiana wetland
habitats. Northeast Gulf Sci. 12:67-72; 1991.
Wilson, E.O. The diversity of life. New York: WW Norton & Co.; 1992.
Wilson, U.W; Atkinson, J.B. Black brant winter and spring-staging use at two Washington
coastal areas in relation to eelgrass abundance. Condor 97:91-98; 1995.
Witkowski, E.T.E The defoliation of woody vegetation by large herbivores at Nylsvley.
Dissertation, Univ. Witwatersrand, Johannesburg, South Africa; 1980.
Wood, R; Gelston, W.L. Preliminary report: the mute swans of Michigan's Grand Traverse
Bay region. Mich. Dept. Nat. Resour. Rep. 2683; 1972.
Wootton, IT.; Oernke, M.P. Latitudinal differences in fish community trophic structure, and
the role of fish herbivory in a Costa Rican stream. Environ. BioI. Fish. 35:311-319;
1992.
Ydenberg, RC.; Prins, H.H. Spring grazing and the manipulation of food quality by
barnacle geese. J. Appl. Ecol. 18:443-453; 1981.
9. Interactions Between Grazing Birds
and Macrophytes
Stuart F. Mitchell and Martin R. Perrow
Introduction
In the past, aquatic birds were largely overlooked by limnologists, receiving scant
attention in hydrobiological journals and no more than passing mention in limnology texts. There has recently been rapid growth in interest in their roles in lake
ecosystems, with the integration of bird studies into intensive limnological programs, comparative investigations over large groups of lakes, an increase in the
number of experimental studies, and increasing contact with water bird biologists
(Kerekes and Pollard, 1994; Farago and Kerekes, 1997). Much of this interest
stems from recent scientific focus on the factors that lead to shallow eutrophic
lakes being dominated alternatively by phytoplankton or by macrophytes (Scheffer et aI., 1993) and management investment in the restoration of eutrophic lakes
to a clear macrophyte-dominated state (e.g., National Research Council [USA],
1992; Broads Authority [UK], 1994; National Environmental Research Institute
[Denmark], 1994). It is now clear that aquatic bird populations may, at times, be
very sensitive to ecological changes in lakes and that they can also play significant
roles in producing such changes.
Our objectives are to review methods for quantifying bird-macrophyte interactions, to discuss examples selected to illustrate some of the interactions that have
been quantified, and to point to areas in which further study may be particularly
fruitful. We emphasize the functional links indicated in Figure 9.1 and do not
consider how the interactions may affect such structural properties as plant or bird
175
176
S.P. Mitchell and M.R. Perrow
Grazing consumption
MACROPHYTE DECLINE -------...
~ ---~
Water
Birds
Senescence
.
Increased - - - - - I Hlotm- ' \
depth
..tion S,d'm,nt jr~
L
Waves
Storms
7P,.nkton
Nutrients
Excretion Defaecation
Figure 9.1. Direct effects and indirect physical and chemical effects of water birds on
macrophytes in relation to some other factors regulating macrophyte abundance in lakes.
Stimulatory effects -+; inhibitory effects -I. (Redrawn from Mitchell and Wass, 1996a,
with kind permission of Elsevier Science-NL, Sara Burgerhartstraat 25, 1055 KV Amsterdam, The Netherlands.)
community diversity or species richness or the potential role of birds in the
dispersal of macrophytes. Although such effects may themselves have functional
consequences, as may feeding interactions among bird species, and food chain
interactions involving carnivorous birds, little is known of them in lakes. We also
neglect the potential effects of the complex food chain interactions that arise from
omnivory in birds, although among aquatic birds only some geese and swans
appear to be strict herbivores (Owen and Black, 1990; Baldassarre and Bolen,
1994). Even those species are not confined to eating aquatic macrophytes, with
geese most typically feeding terrestrially.
Birds directly affect the dynamics of plant biomass by consuming macrophyte
tissue. They may also affect the productivity or rates of change in biomass
indirectly by grazing selectively, damaging the remaining plants, and cycling
nutrients or by a variety of other mechanisms that modify the plants or their
habitats (Fig. 9.1). The effect on the dynamics of the plant community (interaction
strength) depends on grazing consumption, the net effect of these various indirect
positive and negative feedbacks, and the timing of consumption within the plant
growth cycle. These interactions, the ways in which population densities of
aquatic birds respond to changes in trophic status and macrophyte abundance and
9. Grazing Birds-Macrophytes Interaction
177
the role of birds in determining phytoplankton and macrophyte dominance, provide the focus for our study.
Water Bird Abundance and Trophic Status
Indices of trophic status such as phytoplankton (chlorophyll a) and plant nutrient
concentrations are available for many lakes, as is information on aquatic bird
populations. Few attempts have been made, however, to collate large data sets on
aquatic bird popUlation densities and trophic status. Notable exceptions include
the studies of Hoyer and Canfield (1994) on shallow subtropical lakes in Florida
and Suter (1994) on deep Swiss lakes. Hoyer and Canfield estimated annual
average population densities of 50 bird species at 46 lakes of widely varying
trophic status. They found significant correlations of total bird numbers and
biomasses with the trophic indicators total phosphorus (TP), total nitrogen (TN),
and chlorophyll a (r = 0.56-0.61). Nilsson and Nilsson (1978) and Murphy et al.
(1984) have recorded similar correlations of water bird densities to chlorophyll a
and/or TP concentrations in southern Swedish and Alaskan lakes. Suter (1994)
surveyed wintering bird populations over 12 years in 20 generally large (3-580
km2) alpine lakes, in relation to trophic status, in categories defined by TP concentrations. Abundances per unit lake surface area were significantly related to
trophic status only for mallard (Anas platyrhynchos) among the eight herbivorous
or omnivorous species tested and also for the piscivorous cormorant (phalacrocorax carbo). Similar relationships to trophic status have been noted for other
piscivorous birds in lakes in Sweden (Nilsson, 1978), Nova Scotia (Kerekes,
1990), and Alaska (Heglund et al., 1994). Effects of lake morphometry have also
been noted, with population densities sometimes being more closely related to
shore length than to surface area in lakes of similar trophic status (Nilsson, 1978;
Suter, 1994; Kerekes, in press).
Empirical modeling offers considerable promise for predicting water bird densities, although different feeding guilds and different species within guilds are
likely to show different types of response. The consistent positive reponses of
piscivores to trophic status over a wide range of lake depths may be due to fish
biomass generally increasing with increasing eutrophication (Hanson and Peters,
1984) but being largely independent of lake depth (Jeppesen et al., 1997).
Herbivorous and omnivorous species cannot, however, be expected to show
continuous positive responses to increases in TP, as macrophytes, after increasing
progressively with increasing eutrophication (Canfield and Hoyer, 1992), often
then decline abruptly owing to shading by phytoplankton or epiphytes (e.g.,
Scheffer et aI., 1993). Thus in Tomahawk Lagoon, New Zealand, the winter black
swan (Cygnus atratus) population size is related inversely to phytoplankton productivity in the previous summer, owing to the inverse variations of phytoplankton
and perennial macrophytes (McKinnon and Mitchell, 1994). There is some evidence that primary productivity of phytoplankton and macrophytes combined may
be higher in shallow eutrophic lakes during phytoplankton dominance than when
178
S.F. Mitchell and M.R. Perrow
macrophytes predominate (Roijackers, 1985; Mitchell, 1989). It is unclear to what
extent this effect, if it is a general one, may flow up through the food chain to
regulate populations of omnivorous or carnivorous birds. Primary productivity is,
however, likely to be a better predictor of such populations than total phosphorus,
from which it is liable to become uncoupled in shallow, highly eutrophic lakes
(e.g., Mitchell et al., 1988).
Water Birds and Macrophyte Abundance
Among the first to note an effect of macrophyte abundance on water bird populations were Allison and Newton (1972/3), who observed progressive decline in the
populations of mute swan (Cygnus alar), coot (Fulica atra), and pochard (Aythya
ferina) from 1950-1971, in Loch Leven, Scotland, in parallel with a decline in
macrophytes brought about by eutrophication. Dirksen et al. (1991) have similarly
reviewed the wax and wane of Bewick's swan (Cygnus columbianus bewicki)
populations in The Netherlands in relation to the establishment and decline of
freshwater macrophytes after polder construction and subsequent eutrophication.
A violent storm that destroyed macrophyte beds in the large (180 km2 ) Lake
Ellesmere, New Zealand, in 1968 led to a decline in the population of black swans
from 40,000-80,000 in the mid-1960s to 4,000 by 1986 (Williams, 1979; McKinnon and Mitchell, 1994). In Lake Christina, Minnesota, populations of various
dabbling and diving ducks, american coots (Fulica americana), and Canada geese
(Branta canadensis) all declined throughout a progressive lO-year decline in
macrophyte cover and recovered in parallel with the macrophytes in the following
5 years, after a fish removal biomanipulation (Hanson and Butler, 1994). Giles
(1994) showed similar effects of fish removal in a biomanipulated gravel pit by
comparison with an unmanipulated control pit. Similar large, long-term changes in
vegetation and parallel changes in populations of coots, swans, and herbivorous
and omnivorous ducks have been reported, among others, by Hargeby et al.
(1994), Lauridsen et al. (1994), Schutten et al. (1994), van Donk et al. (1994), and
S!Ilndergaard et al. (this volume, Chapter 20). In Hawksbury Lagoon, New Zealand, black swan population densities followed the week-to-week changes in
biomass of aquatic vegetation remarkably closely during a 7-month period (Mitchell and Wass, 1996a). In surveys of groups of lakes, brood densities of black
duck (Anas rubripes) were significantly correlated with macrophyte cover in 32
Nova Scotian lakes (Staicer et al., 1994), and winter black swan populations were
significantly correlated with macrophyte biomass in seven New Zealand lakes
(McKinnon and Mitchell, 1994).
These results suggest that population densities of water birds are often closely
related to the abundance of macrophytes, and further development of general
empirical models for the relationships may prove very useful. Lakes provide many
examples of simple quantitative relationships linking various functional or taxonomic groups. The phosphorus-chlorophyll relationship (e.g., DECO, 1982),
which has proved a powerful tool for lake research and management, is one
9. Grazing Birds-Macrophytes Interaction
179
example, but there are many others (e.g., Hanson and Leggett, 1982; Hanson and
Peters, 1984). Aquatic birds are unlikely to prove an exception, although the generality
of models for birds will be affected by such seasonal factors as territoriality and
migration. For example, coot densities may become related to macrophyte abundance
only in autumn, after territories break up (Perrow et al., 1997).
Until now, most attention has been given to demonstrating the existence of such
relationships in particular lakes. Some of the studies cited above have been
semiquantitative, with only general observations of macrophyte abundance, or are
quantified and presented in ways that do not readily allow comparisons with other
systems. Other studies have shown no clear relationships between aquatic bird
popUlations and macrophyte biomass (Hoyer and Canfield, 1994) or other macrophyte indices (Hoyer and Canfield, 1994; Lillie and Evrard, 1994), and it is
unclear to what extent differences in the results and our present inability to make
general predictions about aquatic bird populations from macrophyte data reflect
the true complexity of the relationships, or whether they res.ult merely from the
paucity of quantitative information expressed in the appropriate units.
What the appropriate units might be is a matter for conjecture. It seems likely,
however, that the closest general relationships involving waterfowl will be revealed by
studies of plant biomass available as food, rather than shoot density, percentage cover,
or other variables such as shore length or the extent of the littoral zone. Macrophytes
at depths beyond the feeding reach of a species are not available to the birds as food,
so that an appropriate unit for water bird population densities might be
Total number of birds on the lake/(ha of lake at depths within feeding reach)
with the available macrophytes being estimated as g/m2 for the area within feeding
reach. Use of these units resulted in better predictions of black swan population
densities than simple numbers and biomasses per unit area (McKinnon and Mitchell, 1994). An alternative unit that may be even more appropriate for tall-growing
macrophytes is g/m 2 in the upper part of the water column that is within feeding reach.
There is scattered evidence of discontinuities in the numerical responses of
some water bird populations at both high and low macrophyte biomasses. Beekman et al. (1991) observed that migratory Bewick's swans abandoned feeding on
Potamogeton tubers and left a lake when they had reduced the tuber density to
about 7 g DW/m2 and coots abandoned feeding on Ruppia at a similar biomass
(Verhoeven, 1980). Black swan numbers on Hawksbury Lagoon show evidence of
a similar discontinuity at filamentous algal biomasses of about 2-3 g DW/m2
(Mitchell and Wass, unpublished data). Black swan populations also tend to
decline when macrophytes or filamentous algae form dense emergent patches that
occupy a large part of the lake surface (e.g., Wass and Mitchell, this volume,
Chapter 18). Use of the algal patches by swans, as indicated by rates of fecal
deposition, was only about 25% of their use of the intervening relatively bare
patches (Mitchell and Wass, unpublished data). This effect might be due simply to
swans finding it difficult to swim through dense plant stands or, alternatively, to
differences in food quality.
180
S.P. Mitchell and M.R. Perrow
An increase in macrophytes that results in more food for many water bird
species may represent a decrease in available habitat or feeding opportunty for
others, and negative relationships between macrophyte cover and bird abundance
have been recorded for several bird species associated with Florida lakes (Hoyer
and Canfield, 1994).
Estimating Food Consumption
If the role of birds in lake ecosystems is to be understood and quantified, it is
essential to know, among other things, how much they eat in nature. Until
adequate estimates become available for rates of food consumption and food
preferences and how they vary among species, seasonally, with age, with food
availability, and with food quality, our understanding of the roles of birds in
aquatic ecosystems will be hampered.
In the absence of direct estimates of food consumption, there has been widespread reliance on bioenergetic estimates that take little or no account of the above
sources of variation or interspecific variation in the birds' anatomy, digestive
physiology, or feeding behavior. Such estimates often have very wide 95% confidence intervals, which may span an order of magnitude (Furness, 1978; Peters,
1983). Bioenergetic modeling essentially involves determining size-based standard metabolic rates (e.g., Zar, 1969; Aschoff and Pobl, 1970) or size-based rates
for metabolism of birds confined to small cages (Kendeigh, 1970), which are then
corrected for the various types of activity in nature, and for the effects of temperature on metabolism (e.g., Woollhead, 1994). These relationships, in turn, have
been used to derive size-energy expenditure relationships in nature (King, 1974).
In attempting to estimate food consumption from modeled energy expenditure, the
already wide confidence intervals are further extended by the need to account for
energy directed to growth, and for differences in the efficiency of assimilation.
Published assimilation efficiencies for waterfowl range from about 20-70%, suggesting that this is a substantial source of additional error. There appear to be systematic
differences in the food consumption of swan and goose species, even after adjustment
for the effects of body size on metabolism, which can be related to differences in their
digestive morphology and feeding behavior (Mitchell and Wass, 1995).
Energy expenditure by individual birds in nature may be estimated by determining the rates of elimination of experimentally administered doses of isotopical1y enriched (double labeled) water (Nagy, 1987). This method, used in conjunction
with a knowledge of the assimilability of the food, has strong possibilities for
estimating food consumption by waterfowl. The less direct bioenergetic methods
offer no more than crude approximations, and are perhaps best avoided in most
circumstances. I
I Although crude, such estimates are greatly preferable to scaling for body weight by
simple proportions, which can lead to large errors. Metabolism normally varies with body
weight to the power of about 0.65-0.75 (e.g., Bertalanffy, 1968).
9. Grazing Birds-Macrophytes Interaction
181
There is a wide range of alternative methods available. Geese and ducks lend
themselves readily to experimental studies of food consumption in a variety of
conditions ranging from small cages to near-natural conditions (e.g., Marriott and
Forbes, 1970; Mattocks, 1971; Gere and Andrikovics, 1994), and controlled feeding experiments have also been done with captive mute swans (Mathiasson, 1973)
and coots (Verhoeven, 1980).
Food consumption may also be calculated by measuring rates of deposition of
feces and analyzing the food and feces for cellulose, on the assumption that it is
not digested (Ebbinge et aI., 1975; Cargill and Jeffries, 1984; Mitchell and Wass,
1995). Any digestion of cellulose will lead to food consumption being underestimated. In food mass balance studies with geese, Buchsbaum et al. (1986) report
that up to 28% of cellulose was digested, but Marriot and Forbes (1970) and
Sedinger et al. (1989) found no evidence of cellulose digestion, and Mattocks
(1971) could detect no cellulase activity in the gut ceca.
Beekman et al. (1991) estimated grazing consumption by Bewick's swans by
measuring rates of decline of dormant tubers of Potamogeton in naturally grazed
areas and ungrazed control areas, and relating the difference to the abundance of
the birds. They note, however, that other species may at times accompany the
swans and compete for tubers that they have uprooted. When birds do not feed
continuously, it may be possible to identify recently ingested material in the gut
and thereby estimate short-term rates of consumption (e.g., Gauthier, 1993).
Perrow et al. (1997) estimated food consumption by coots from direct observations, by relating the length of food items to the length of the bill, and by
determining the biomass per unit bill length for various macrophyte species and
other food items. Most waterfowl species, however, handle food items in ways that
preclude this method.
Little is also known of the extent and causes of food selectivity among waterfowl. The food species eaten by water birds often change seasonally (e.g., Black
and Rees, 1984; Owen and Black, 1990; Baldassarre and Bolen, 1994; Grant et al.,
1994). This may often be no more than a response to changing availability (Perrow
et aI., 1997), but ontogenetic shifts occur in ducks (Baldassarre and Bolen, 1994)
and in coots, in which young are fed predominantly on high-protein insects before
they become self-sufficient herbivores or omnivores (Perrow et aI., 1997). Female
ducks increase both their food intake and the proportion of animals in the diet
during laying (Noyes and Jarvis, 1985) and the diet of adult coots shows similar
changes during parental care (Perrow et aI., 1997). Food preferences might be
related to protein content (Rees, 1990), texture, secondary plant metabolites, or
other chemical constituents (cf. Lodge et aI., this volume, Chapter 8; Cronin, this
volume, Chapter 21). McKinnon (1989) found no indication of any selectivity
between charophytes and angiosperms among black swans on four lakes, but the
results were highly variable, and the discriminatory power of the indices used
was low. The relative abundances of different stable isotopes of carbon and
nitrogen in the tissues of birds and their potential food species provide promising opportunities for unraveling complex food linkages (Hoyer et aI., this
volume, Chapter 23).
182
S.F. Mitchell and M.R. Perrow
Little is also known of individual variations in food consumption in nature. It
must, however, be a cause for concern when estimates are based on small numbers
of individual birds. For example, Marriott and Forbes (1970) found an almost
fourfold range of variation in daily food intake by captive Cape Barren geese fed
ad libidum. Similarly, functional feeding responses have not been investigated for
natural water bird populations. Food consumption varies with food availability in
many animals, but such responses might be relatively unimportant in water birds,
first because of their strong numerical responses-if food levels become suboptimal they may simply move elsewhere. Second, feeding occupies only a small
part of the day in many species (Black and Rees, 1984; Baldassarre and Bolen,
1994), and additional time might be devoted to it when food is in short supply, to
maintain a constant daily ration.
Given the difficulties of determining rates of food consumption in nature,
authors, having obtained an estimate, have generally been content to regard it as a
constant. Energy requirements are, however, known to vary with temperature, and
phases of the molt and reproductive cycles, and the extent to which these variations are translated into differing rates of food consumption or accommodated by
changes in body fat requires further study.
Water Birds as Consumers
Even at their most abundant, birds appear to be a minor component of the biomass
of lakes. Bird biomasses at 46 shallow lakes in subtropical Florida ranged from
0.001 to 0.47 g/m2 annual average live weight (0.01-4.7 kg/ha) (Hoyer and
Canfield, 1994). The mean was only 0.11 g/m2 (ca 0.04 g DW/m2) by comparison
with the mean wet weight biomass of submerged and floating leaved macrophytes
of 3,100 g/m2.
From cooler regions, Poysa (1983) reports a figure equivalent to 0.16 g/m21ive
weight for a shallow, eutrophic, Finnish lake. In Hawksbury Lagoon (46 S), the
black swan density becomes as high as 25 birds/ha, and the average of 10 birds/ha
during long phases of macrophyte dominance corresponds to 1.9 g DW/m2
(5.6 g/m2 or 56 kg/ha live weight), which is higher than for any of the Florida lakes
cited and again very low in relation to the maximum macrophyte (macroalgal)
biomass of 200 g DW/m2. These figures may be near the upper limit for annual
average water bird biomasses (Mitchell and Wass, 1995), although flamingos
(Hurlbert and Chang, 1983), some roosting populations, and populations that are
artificially fed may be exceptions. Bird biomasses may be seasonally three to four
times higher than this on lakes where terrestrially feeding migratory geese gather
(from Manny et aI., 1994). Bird biomasses also appear to be low in relation to
those commonly observed for phytoplankton and benthic fauna (frequently> 109
DWI m2), and lower than those often recorded for fish (100-400 kg live weight/ha
or ca 3-12 g DW/m2) and zooplankton (0.1-0.5 g DW/m2) (data from various
sources). These figures suggest a rather minor role for birds as consumers in lake
ecosystems. As they are among the largest of the lake fauna, the role of birds in
0
9. Grazing Birds-Macrophytes Interaction
183
biological productivity will be further reduced by inverse size/metabolism relationships but increased by the additional energy demands of endothermy.
Ki~rboe (1980), who used bioenergetic estimates of food consumption in
Ringk~bing Fjord, Denmark, suggested limits for consumption of macrophytes by
birds of between 15-60% of the annual productivity but also cautioned about the
use of this method. Woollhead (1994), using similar methods, suggests a figure of
13% for eutrophic Lake Esrom, Denmark. Using more direct methods, Clausen
(1994) obtained a figure of 12% (range, 8-21 %) for Brent geese (Branta bemicla)
grazing on Zostera, and black swan consumption of macroalgal productivity at
Hawksbury Lagoon was 12 and 16% during different years (Mitchell and Wass,
1996a; Wass and Mitchell, this volume, Chapter 18). Although many of the 19
figures cited by Cyr and Pace (1993) are substantially higher than these, those
estimates do not represent consumption (Mitchell and Wass, 1996b). Figures for
total annual consumption expressed as a fraction of maximum plant biomass also
overstate the fraction of net primary productivity consumed, as consumption
during biomass increase is a component of the productivity, which should be
added to the net biomass increase.
Little information is available on consumption by birds in relation to that by
other herbivores. In Lake Zwernlust, The Netherlands, calculated annual consumption of macrophytes by rudd (Scardinius erythrophthalamus) varied only
from 170 to 360 kg, from 1990 to 1996, whereas consumption by coots varied
from 30 to 1,200 kg as the coot popUlation changed in tandem with variations in
the macrophyte biomass (van Donk, this volume, Chapter 19).
Their low biomasses and generally small role as consumers suggest that possibilities for birds to have major effects on the dynamics of plant biomass may be
limited to temperate waters in periods in which macrophyte productivity is loweither early in the growth phase of seasonal species, when biomass is low, or in
autumn, when seasonal plant growth has slowed or ceased. Alternatively, it requires that the net indirect effect of birds on the plants should be negative and large
in relation to the direct effect of consumption and/or that the timing of grazing
should be important.
Considerations of the Timing of Grazing
Consumption of plants removes not only plant material but also the future productive potential of that material. Consumption of tissue during active growth affects
rates of plant biomass increase, but grazing after seasonal growth has ended is
inconsequential, at least for that season's productivity (Ki~rboe, 1980). For this
reason alone, biomass consumption may give misleading indications of interaction
strength.
Because of the interdependence of biomass and productivity, the effect of
removing growing tissue is nonlinear. If the productivitylbiomass quotient (PIB)
remains constant, the effect will be exponential, and the effects of earlier consumption will become disproportionately larger as time progresses. This effect
184
S.F. Mitchell and M.R. Perrow
3
4
I
I
a
~I..-
I
t
t
2
_ .....
--:::==F:;;::::===
TIME
Figure 9.2. Generalized time course of plant biomass increase to equilibrium biomass in
herbivore exclusion experiments to illustrate the effects of exponential growth (period a)
and density limitation (period b) on the results for experiments with two different starting
times (l and 2) and two alternative finishing times (3 and 4). x, y, ungrazed biomasses; z,
grazed biomasses.
("biomass compounding") may cause even a small amount of consumption early
in the growth cycle to have a large effect (Fig. 9.2). The ultimate effect of grazing
on biomass will be insignificant, however, if the plant community can attain the
carrying capacity of the system despite being grazed. In the absence of grazing, it
would merely reach it sooner. Approach to the carrying capacity is marked by
declining PIB and declining rates of biomass increase to produce the familiar
sigmoid seasonal biomass curve (Fig. 9.2).
Thus even in systems with identical grazing rates and plant growth rates, the
outcomes may be quite different, ranging from a very large effect on biomass to
none at all. The significance of exponential growth is illustrated by the study of
Wass and Mitchell (this volume, Chapter 18), in which the biomass of experimentally ungrazed macroalgae became five times higher than that in the natural grazed
system, although grazing consumption was only 12% of the net plant productivity.
The major effect of grazing arose not from the biomass consumed but from the
loss of the future growth potential of that material. That material was only ever
represented as potential, and it was never present in the real grazed system to be
either eaten (cf. Cyr and Pace, 1993) or dislodged by the herbivores (cf. Cattaneo
and Mousseau, 1995). The importance of density limitation is illustrated by the
9. Grazing Birds-Macrophytes Interaction
185
fact that growth of the natural community continued after the experimentally
protected plants had become density limited, and the ultimate biomass difference
was effectively zero. The early occurrence of density limitation in the ungrazed
plants was also an experimental artifact, in the sense that it occurred prematurely,
long before any such effect might have been detected in the real grazed system (cf.
Fig. 9.2). The effect of grazing on biomass is therefore highly time dependent and
highly dependent on the plant growth cycle (Mitchell and Wass, 1996b).
The problems posed by biomass compounding and density limitation can be
solved by assessing grazing, not in terms of biomass or ultimate biomass outcomes
but as a dynamic process affecting rates of change in biomass, and with proper
regard to the growth cycle of the plants. These objectives can be achieved by
exponential modeling, with growth, interaction strength, and grazing all being
expressed as plant tissue-specific rates. 2 These effects have important implications for the ways in which herbivore exclusion experiments are conducted and
interpreted (see below).
At a different level, grazing consumption may show large seasonal variations.
With the arrival of migratory flocks at many north temperate waters in autumn,
grazing impacts increase dramatically, and water birds may contribute substantially to the decline in plant biomass after seasonal growth has ended (Anderson and
Low, 1976; Ki~rboe, 1980; Esler, 1989). Conversely, during the breeding season
many species, including coots and most swan species, are territorial, which reduces grazing impacts (Perrow et al., 1997).
Indirect Feedback Effects of Birds on Macrophytes
Nutrient Inputs and Cycling
The small role of birds in the biomass of lakes suggests that their body wastes will
also play only a minor part in nutrient dynamics, and that appears to be normally
true. Marion et al. (1994) report contributions by bird feces to annual loads of 0.4
and 0.7% of TN and 2.4 and 6.6% of TP for two study periods on shallow Lake
Grand-Lieu, France. Nutrients in mallard feces are equivalent to less than 1% of
the external TP entering Lake Kis-Balaton and 2% of the TN (from Gere and
Andrikovics, 1994). Fecal inputs of phosphorus to Hawksbury Lagoon by a dense
black swan population (lO/ha) were sufficient to generate concentrations of 1530 mg/m 3 on the DECO (1982) loading-concentration model (Mitchell and Wass,
1995) or about 5-10 mg/m3 when only the soluble P is considered. These con2The model may at times be fitted to seasonal biomass data. At other times, the essentially
exponential nature of growth may be obscured by changing environmental conditions
(including changes in grazing rates), the interplay between aboveground and underground
components of the biomass (the latter often neglected), and progressive germination,
coupled with high sampling variation (see, e.g., data of Jupp and Spence 1977; S~ndergaard
et aI., 1996). It may still be fitted to individual sampling intervals during continuous growth,
however.
186
S.F. Mitchell and M.R. Perrow
centrations, although significant by the standards of many lakes, were small in
relation to the annual average TP concentration of 340 mg/m 3. N inputs were
similarly small, and weekly fecal loadings of inorganic N and soluble reactive P
were less than 1% of the maximum observed weekly increases in concentrations.
Other studies, based on bioenergetic estimates or proportionately scaled body weights,
tend to support these findings. For example, the annual contribution by bird feces
(much of it as an internal load) averaged only 2.4% of the annual external phosphorus
load in 14 Florida lakes and 6% of the phosphorus present in the water and macrophytes in a larger sample of 46 lakes (Hoyer and Canfield, 1994).
There are, however, striking exceptions. Large overwintering flocks of terrestrially feeding Canada geese contributed 70% and 27%, respectively, of the
annual TP and TN loads to Wintergreen Lake, Michigan, and the lake's very dense
phytoplankton populations are attributed to their influence (Manny et aI., 1994).
The eutrophication and loss of macrophytes in Hickling Broad, England (120 ha),
are also attributed to nutrient inputs by flocks of roosting gulls that have reached
abundances of 250,000 (Leah et aI., 1978; Moss and Leah, 1982). Resident bird
populations may be unimportant, but that is certainly not always true for roosting
flocks and perhaps also for the high populations that are sometimes sustained by
artificial feeding. Furthermore, the inputs from birds that feed elsewhere represent
an external load, rather than a recycling of nutrients already present in the system.
These studies, the rather large fractions of total Nand P that are present in
soluble forms (Mitchell and Wass, 1995), and the apparently rapid mobilization of
sedimented faecal nutrients (Moss and Leah, 1982) all support the view that fecal
deposition in lakes benefits principally the phytoplankton (and possibly epiphytes), which compete with macrophytes for light, and is therefore potentially
harmful to macrophytes in such guano trophic systems.
Relief of Density Suppression of Growth
Plants may become density limited through a variety of mechanisms such as resource
depletion or self-shading (e.g., Lodge, 1991). Grazing at these times may increase
tissue-specific growth rates. This effect has been demonstrated for water lilies grazed
by invertebrates (Wallace and O'Hop, 1985) and for salt marsh vegetation grazed by
geese (Cargill and Jeffries, 1984). It is unlikely, however, that the continual replacement of photosynthetic tissue could be achieved without cost to the underground
reserves, which were not considered in either study. In the salt marsh, the underground
biomass vastly exceeded the aboveground biomass, and it is unlikely that any significant increase in plant productivity would have been demonstrated. Daily grazing
consumption by even large waterfowl populations is only a minute fraction of the
macrophyte biomass when plant biomass is at the carrying capacity and density
limited (see above), and this effect is unlikely to be significant.
Plant Damage and Selective Grazing
There are various published references to grazing water birds being wasteful
feeders that consume only a portion of what they break off or uproot (e.g.,
9. Grazing Birds-Macrophytes Interaction
187
Berglund et al., 1963; Owen and Black, 1990). Selective grazing on the most
actively growing tissues will similarly have a greater effect on the plants than
would be predicted from a simple consideration of the amount of tissue consumed.
These effects are widely postulated to be important (e.g., Lodge, 1991), but they
have not been quantified. Filamentous algae, charophytes, and aquatic angiosperms can be expected to differ in their sensitivities to these effects, owing to
differences in their morphological complexity. Thus to an algal filament, each cell
is worth as much as any other, but angiosperms must maintain a high level of
morphological integrity, and charophytes can be expected to occupy an intermediate position.
Physical Effects
Bioturbation by birds swimming, feeding, or wading in shallow water might affect
the light climate of macrophytes directly and also, indirectly, by increasing the rate
of nutrient supply from the sediment to phytoplankton (cf. S(ljndergaard et al.,
1992). Mitchell and Wass (1996a) concluded that direct optical effects from black
swans in a very shallow, wind-swept 25-ha lake were insignificant in relation to
those of sediment resuspension by waves and high phytoplankton concentrations,
but they are potentially more important in smaller water bodies where wave
resuspension is less frequent or where dense populations of waders gather (Comin
and Herrera, 1997).
Effects of Birds on Macrophyte Dynamics: Interaction Strengths
Interaction strengths can be determined from controlled herbivore exclusion experiments. In studies using this method, attempts to overcome the problem of high
sampling variability and difficulties associated with the need for destructive sampling to estimate biomass have led to use of a wide variety of exclosure types and
numbers, sampling intervals, experimental periods, and macrophyte parameters
measured, making comparisons among studies often difficult. Comparison of
popUlation density- or species-specific impacts is further inhibited by the paucity
of information on bird use of the grazed control areas, and the concern has been
largely with (biomass) outcomes, rather than the dynamic processes leading to
them, which also presents difficulties of interpretation.
Biomass effects are sensitive to when experiments both begin and end in
relation to plant growth cycles. A later start will produce a smaller difference in
biomass between the grazed and ungrazed communities after any time interval up
to equilibrium (Fig. 9.2) and may also affect the ultimate biomass outcome (i.e.,
the later the start to herbivore exclusion, the later the carrying capacity will be
reached and the greater the probability that seasonal or other factors will intervene
to stop growth before it is attained). Similarly, for experiments in which herbivores
are added rather than excluded or when transplants are used, the longer the plants
are grown before the addition, the less will be the apparent effect after any given
188
S.P. Mitchell and M.R. Perrow
time interval. Experiments in which only initial and final biomasses are presented
allow neither interaction strengths nor ultimate biomass outcomes to be estimated,
owing to the time course of the effect on biomass being nonlinear (Fig. 9.2).
Ultimate biomass outcomes are important, particularly for lake management,
although it must be recognized that they reflect not just herbivory but the interactions of herbivory, carrying capacity, growth cycles, and the starting time of
experiments. They are likely to be highly lake- and experiment-specific. Published
figures indicate a range of effects from near zero (Anderson and Low, 1976;
Ki0rboe, 1980; Perrow et aI., 1997; Wass and Mitchell, this volume, Chapter 18),
to almost 90% (Van Wijk, 1988; Lauridsen et aI., 1993).
The problems of expressing herbivore impacts adequately in units of biomass
are intractable-a static concept cannot be used to express a dynamic effect.
Exponential modeling (or the less attractive logistic alternative) solves these
problems, and its difficulties are minor by comparison. Further advantages are that
the potentially large effects of any small initial difference in biomass between
control and experimental plots are removed and that with the units being additive,
the effects of grazing on plant growth can be compared directly with those of other
factors such as light.
Interaction strength can be expressed as the difference between the instantaneous growth rates (g plant/g plant/day) of the ungrazed (Bill) and grazed (Big)
plants. The relative effect on the plants (relative interaction strength) might be
expressed as Big I Bill. The concept of a per capita interaction strength is no longer
appropriate, as the effect becomes related to population density rather than population size. We shall use the term specific interaction strength, which can be
expressed as the difference between the instantaneous growth rates of the ungrazed and grazed plants per unit herbivore, or
where N is the population density of the herbivore (no./ha or biomass/ha). To
obtain this information requires no more than well-designed herbivore exclusion experiments, adequate monitoring of the birds, and use of the exponential model. Coupled with estimates of grazing consumption, it may also allow
the direct effects of herbivory to be isolated from the indirect (Mitchell and
Wass, 1996b). With herbivory expressed in these units, it may be possible to
make better comparisons among ecosystems and among different plant and
herbivore species.
A reanalysis of some exclosure studies in which time course data were
provided and which fit closely to a seasonal exponential model indicates that
water bird grazing may have substantial dynamic effects on aquatic plant
communities, with instantaneous rates of biomass increase in the grazed plants
being as low as 29% of those for the ungrazed plants in experiments with
transplants in pots (where the relatively high plant biomasses may have attracted increased grazing attention from the birds) or 50% in natural plant
communties (Table 9.1).
189
9. Grazing Birds-Macrophytes Interaction
Table 9.1. Interaction Strengths (Biu - Big) and Relative Interaction Strengths (Big/Biu)
Calculated from Published Figures for Waterfowl Grazing on Submerged Macrophytes
during Spring-Summer Growtha
Lake
Lauwersmeet'
Vreng a
C
Grazers
Biu
Big
(B iu - Big)
Big/Biu
Coots, etc.
Coots
0.035
0.045
0.048
0.083
0.071
0.044
0.020
0.020
0.014
0.064
0.059
0.027
0.055--0.089
0.027--0.029
0.015
0.025
0.034
0.019
0.021
0.017
0.007
0.008-0.028
0.57
0.44
0.29
0.77
0.83
0.63
0.89--0.93
0.49--0.78
b
c
d
Stigsho1md
Hawksburye
HawksburY
Coots
Swans
Swans
0.037--0.055
aBiu, Big, exponential growth rates of un grazed and grazed plants, respectively (per day).
b Data from Verhoeven, 1980.
cData from Lauridsen et aI., 1993; transplants in pots; derived from total shoot length per pot, which
correlated closely with biomass; a, sheltered sites, mud substrate; b, sheltered sites, sand substrate;
c, exposed sites, mud substrate; d, exposed sites, sand substrate.
dData from S!!lndergaard et a!., 1996.
eData from Mitchell and Wass, 1996a; calculated from food consumption estimate.
fWass and Mitchell, this volume, Chapter 18, spring-summer.
Interactions and Threshold Effects: Grazing and Stable States
Aquatic plant production may at different times (e.g., seasonally) lie either above
or below the grazing consumption. When plants are close to this threshold, and
without the buffer represented by accumulated biomass reserves, even a slight
change in grazing pressure by birds may have major effects on the ecosystem. On
one hand, it may lead to complete suppression of increase in the plants, on the
other to escape from grazing control and ultimate growth to the maximum biomass
that the system can sustain. These different outcomes have been observed within
the same lake (Wass and Mitchell, this volume, Chapter 18), though such fine
balances may not be common or persist for long, as at the typical spring-summer
growth rates of macrophytes and filamentous algae, the biomass neccessary for
escape from complete suppression by grazing may be only a few grams dry weight
per square meter.
Even if grazing birds do not by themselves tip the balance of a lake in favor of
phytoplankton dominance, they may do so in conjunction with other factors such
as low benthic illuminance. Coots, and even swans, may graze to depths greater
than the euphotic depth and have the potential to maintain macrophytes in a
permanently light-starved condition. Anything that delays seasonal macrophyte
growth or recovery after biomanipulation allows more time for the occurrence of
stochastic events that favor phytoplankton dominance, such as resuspension of
lake sediments by storms.
190
S.F. Mitchell and M.R. Perrow
There have been few investigations of bird grazing in relation to other factors
affecting macrophyte dynamics, but in Lake Vreng, Denmark, the effect of coots
grazing on macrophyte transplants grown in pots was similar in magnitude to that
of exposure to waves and greater than the effect of substrate type (Table 9.1;
Lauridsen et al., 1993). In Hawksbury Lagoon, grazing effects were small in
relation to those of variations in the benthic light climate produced by fluctuations
in phytoplankton and sediment resuspension (Mitchell and Wass, 1996a). Nor are
birds the only grazers. The calculated relative consumption by rudd and coots
varies widely in Lake Zwemlust, and their combined effect may have contributed
to a decline in macrophytes and a change in the species composition (van Donk
and Otte, 1996). There is also evidence that some water bird populations may be
greatly affected by the abundance of fish that consume the same food species
(Giles, 1994). The unraveling of such food web complexities and their relation to
stable states offers fascinating problems at a variety of scales (see other chapters,
this volume).
The question of whether the effects of heavy grazing by birds on the overwintering stages of plants in one year carryover to reduce plant productivity in the
next year is of considerable interest for the long-term stability of the macrophytedominated state (S0ndergaard et aI., 1996; van Donk and Otte, 1996; Perrow et aI.,
1997). Beekman et al. (1991) showed that swans abandoned grazing on tubers of
Potamogeton before they reduced the tuber density sufficiently to affect the next
year's plant productivity, and Clausen (1994) obtained similar results for Zostera
grazed by brent geese. Anderson and Low (1976), however, demonstated that
water bird grazing on Potamogeton tubers in one year could have substantial,
albeit local, effects on biomass development in the following year. Apart from this
potential effect, which requires more study, the main role of birds in promoting
phytoplankton dominance may lie in complementing other factors that also inhibit
macrophyte growth.
Conclusions
Aquatic bird populations are often strongly affected by changes in macrophyte
abundance and changes in the trophic status of lakes. It would be very useful to
quantify these relationships, produce predictive models, and define the limits of
those models. Such a framework may allow outliers to be recognized and permit
the influence of both food and other factors to be better evaluated.
The biomasses of birds appear to be normally small in relation to those attained
by phytoplankton, macrophytes, benthic invertebrates, fish, and zooplankton. As
would therefore be expected, birds typically contribute little to nutrient budgets
and consume only a small fraction of the annual macrophyte production. These
simple facts belie their sometimes substantial roles in regulating the rates at which
macrophyte biomass increases during spring-summer growth and the ultimate
biomass attained. Their major influence appears to derive not from what they
consume but from the loss of the future growth potential of that material, com-
9. Grazing Birds-Macrophytes Interaction
191
pounded through the growth period. In angiosperm communities in particular, this
negative effect on the plants will be reinforced by plant damage and wastage
during feeding. The positive feedback effect of birds due to reduction in self-shading, which may be important for tissue-specific plant growth rates in other systems, is likely to be insignificant. Nutrient recycling, a positive feedback in other
systems, becomes negative for macrophytes in lakes but is, in any case, significant
only for some lakes where birds that feed elsewhere use the lake for roosting. The
net indirect feedback effect of birds on macrophytes is thus almost certain to be
negative. Although this effect has not been quantified, a theoretical and practical
framework exists that would allow it to be done.
In theory, birds may by themselves induce a loss of macrophytes and cause
lakes to switch to phytoplankton dominance. The requirement is simply that the
interaction strength should exceed the true plant growth rate. This occurs when
plants stop growing in autumn, but at that time there is often a large accumulated
biomass, including seeds and other resting stages, to provide protection from
extinction. The plants' survival may also be aided by changes in bird feeding
behavior at low threshold biomasses. The extent to which heavy grazing on the
overwintering stages of macrophytes may influence plant production in the following spring and summer remains unclear and requires study in a range of
different lakes. Interaction strengths can also exceed the true growth rates of
actively growing plants, at least briefly, but this situation may not be common.
Tropical systems with persistently high macrophyte biomasses presumably lack
this vulnerability to grazing by birds.
To better understand and quantify these interactions, there is a need to quantify
food consumption by water birds in nature and to determine how it varies with
size, age, season, species, food availability, and food quality and to relate this
information to the biomasses and growth rates of available macrophytes. There is
a need to isolate the feedback effects of the birds on the plants, to quantify relati ve
and specific interaction strengths, to compare these features among lakes and
among plant and bird species, and to produce, test, and refine predictive models
for these effects. There is a need for further studies in which bird herbivory is
investigated, not in isolation but as part of integrated studies on other factors also
affecting macrophyte dynamics. Multifactor experiments may answer the important question of the extent to which grazing effects are additive with other factors
influencing plant growth rates or interactive with them.
To achieve these objectives, limnologists must recognize the unique features of
birds among the lake fauna and adapt their methods accordingly. The macrophyte
biomass parameters that are relevant to zooplankton or fish may not always be so
for birds, and the weekly or two-weekly sampling schedules that we commonly
use for other things might be quite inappropriate for animals whose use of different
regions of a lake may vary by the hour.
The task of producing quantitative, predictive models for macrophyte-bird
interactions should not be underestimated. Apart from the complexities of the
relation of food consumption to interaction strengths, food consumption per bird
may vary diurnally, seasonally, with age, and with other variables. Birds' use of
192
S.P. Mitchell and M.R. Perrow
different regions of a lake may vary on short time scales, and human disturbance
of birds may obscure relationships that would otherwise be significant. The plants
are frequently patchy. They also often vary regionally within lakes, and they are
difficult to sample quantitatively. Not all of them may be accessible, because of
the the birds' restricted feeding depths. Even without the massive complexities of
omnivory, the birds may complicate matters further by feeding on marginal emergent plants or terrestrial vegetation, as well as submerged plants, by seasonal or
local migration, aggregation, territoriality, or other complex behaviors that may
influence both their effect on the plant community and their responses to changes
in it.
If macrophyte-bird interactions represent a challenge, then they also present
opportunities that may not be readily available in other plant-herbivore systems.
The dramatic losses of aquatic macrophytes from lakes, relating to eutrophication,
and their recovery through human intervention or natural processes can be expected to have equally large effects on herbivorous bird populations. Birds' mobility means that the option of emigration to seek a better habitat is always
available, and recruitment is not restricted to a particular brief breeding season as
in many other vertebrate herbivores. Time lags and imbalances between populations and their food resources, which hamper simple analysis in many other
systems (e.g., Carpenter et aI., 1985), may be reduced, and with populations being
closer to equilibria, simple relationships are more likely to emerge. The hypothesis
that bottom-up effects of macrophytes on aquatic birds are relatively strong should
therefore be readily testable, with opportunities to use neighboring biomanipulated and unmanipulated lakes as controls.
The converse hypothesis, that grazing birds have large effects on the structure
and dynamics of aquatic plant communities, offers similar opportunities for testing
on natural systems. The sudden arrival of large migratory flocks of birds on many
northern waters facilitates "before-and-after" studies of their effects, again with
opportunities for replication at the ecosystem level by manipulations to inhibit
bird use. Stands of aquatic plants, often essentially monospecific, which may be
grazed by a bird as the only large herbivore species, offer a simplicity that is rare
in plant-large herbivore systems, with a wide range of possibilities for field
enclosure/exc1osure experiments. The different sensitivities of filamentous algae,
charophytes, and angiosperms to different indirect feedback effects of herbivores
offer unique opportunities for isolating feedbacks from each other and from the
primary effect of consumption. Macrophyte-bird interactions present wide opportunities for formulating and testing hypotheses that may be significant not only for
lake ecology but for ecology in general.
Acknowledgments. We are very grateful to the Danish National Environmental
Research Instititute for their invitation to attend the workshop and for financial
assistance, to the Workshop Organising Committee for their splendid efforts, and
to other colleagues who attended for much stimulating discussion. We also thank
Mark Hoyer, Joe Kerekes, David Lodge, Torben Lauridsen, and Peter Webb for
their comments on this chapter. Rob Wass kindly redrew the figures.
9. Grazing Birds-Macrophytes Interaction
193
References
Allison, A.; Newton, I. Waterfowl at Loch Leven, Kinross. Proc. R Soc. Edinb. (B), 74
(24):365-379; 1972n3.
Anderson, M.G.; Low, J.B. Use of sago pondweed by waterfowl on the Delta Marsh,
Manitoba. J. Wild!. Manage. 40:233-242; 1976.
Aschoff, J.; Pohl, H. Der Ruheumsatz von Vogeln als Funktion der Tageszeit und der
Korpergrosse. J. Ornithol. III :38-47; 1970.
Baldassare, G.A.; Bolen, E.G. Waterfowl ecology and management. New York: Wiley;
1994.
Beekman, 1.H.; van Eerden, M.R.; Dirksen, S. Bewick's swans Cygnus columbianus
bewickii utilising the changing resource of Potamogeton pectinatus during autumn in
The Netherlands. Wildfowl (Supp!. 1):238-248; 1991.
Berglund, B.E.; Curry-Lindahl, K.; Luther, H.; Olsson, Y.; Rodhe, W.; Sellerberg, G.
Ecological studies on the mute swan (Cygnus alar) in southeastern Sweden. Acta
Vertebratica 2; 1963.
Black, 1.M.; Rees, E.C. The structure and behaviour of the whooper swan population
wintering at Caerlaverock, Dumphries and Galloway, Scotland; an introductory study.
Wildfowl 35:21-36; 1984.
Broads Authority (UK). The Broads plan-no easy answers. Norwich, England: Broads
Authority Consultation Document, Broads Authority; 1994.
Buchsbaum, R; Wilson, J.; Valiela, I. Digestibility of plant constituents by Canada geeses
and Atlantic brant. Ecology 67:386-393; 1986.
Canfield, D.E., lr.; Hoyer, M.Y. Aquatic macrophytes and their relation to the limnology of
Florida lakes. SP115. Institute Food and Agricultural Sciences, University of Florida;
1992.
Cargill, S.M.; Jefferies, RL The effects of grazing by lesser snow geese on the vegetation
of a subarctic salt marsh. 1. App!. Eco!. 21:669-686; 1984.
Carpenter, S.R.; Kitchell, 1.E; Hodgson, 1.R Cascading trophic interactions and lake
productivity. BioScience 35:634-638; 1985.
Cattaneo, A.; Mousseau, B. Empirical analysis of the removal rate of periphyton by grazers.
Oecologia 103:249-254; 1995.
Clausen, P. Waterfowl as primary consumers in shallow water fiord areas. PhD thesis,
National Environment Research Institute/University of Aarhus, Kal¢, Denmark; 1994.
Comin, F.A.; Herrera, 1.A. Enhanced release of nutrients from the sediment to the water
column by flamingos' footprints. In: Farag6, S.; Kerekes, 1., eds. Workshop Proceedings. Limnology and Waterfowl, Monitoring, Modelling and Management. Working
Group on Aquatic Birds, Societas Internationalis Limnologiae and International Waterfowl and Wetlands Research Bureau. Sarr6d/Sopron, Hungary, 21-23 November 1994.
Wetlands Int. Pub!. 43; 1997.
Cyr, H.; Pace, M.L. Magnitude and patterns of herbivory in aquatic and terrestrial ecosystems. Nature 361:148-150; 1993.
Dirksen, S.; Beekman, 1.H.; Slagboom, T.H. Bewick's swans Cygnus columbianus bewickii
in The Netherlands: numbers, distribution and food choice during the wintering season.
Wildfowl (Supp!. 1):228-237; 1991.
Ebbinge, B.; Canters, K.; Drent, R. Foraging routines and estimated daily food intake in
barnacle geese wintering in the northern Netherlands. Waterfowl 26:5-19; 1975.
Esler, D. An assessment of American coot herbivory of Hydrilla. .T. Wild\. Manage. 53:
1147-1149; 1989.
Farag6, S.; Kerekes, 1., cds. Workshop Proceedings. Limnology and Waterfowl, Monitoring, Modelling and Management. Working Group on Aquatic Birds, Societas Internationalis Limnologiae and International Waterfowl and Wetlands Research Bureau.
Sarr6d/Sopron, Hungary, 21-23 November 1994. Wetlands Int. Pub!. 43; 1997.
194
S.E Mitchell and M.R. Perrow
Furness, R.W. Energy requirements of seabird communities: a bioenergetics model. 1.
Anim. Ecol. 47:39-53; 1978.
Gauthier, G. Feeding ecology of nesting snow geese. J. Wildl. Manage. 57:216-223;1993.
Gere, G.; Andrikovics, S. Feeding of ducks and their effects on water quality. Hydrobiologia 279/280:157-161; 1994.
Giles, N. Tufted duck (Atyhyafuligula) habitat use and brood survival increases after fish
removal from gravel pit lakes. Hydrobiologia 279/280:387-392; 1994.
Grant, T.A.; Henson, P.; Cooper, J.A. Feeding ecology of trumpeter swans breeding in south
central Alaska. J. Wildl. Manage. 58:774--780; 1994.
Hanson, J.M.; Leggett, W.C. Empirical prediction of fish biomass and yield. Can. J. Fish.
Aquat. Sci. 39:257-263; 1982.
Hanson, J.M.; Peters, R.H. Empirical prediction of crustacean biomass and profundal
macrobenthos biomass in lakes. Can. J. Fish. Aquat. Sci. 41:439--445; 1984.
Hanson, M.A.; Butler, M.G. Responses to food web manipulation in a shallow waterfowl
lake. Hydrobiologia 279/280:457--466; 1994.
Hargeby, A.; Andersson, G.; Blindow, 1.; Johansson, S. Trophic web structure in a shallow
eutrophic lake during a dominance shift from phytoplankton to submerged macrophytes.
Hydrobiologia 2791280:83-90; 1994.
Heglund, P.J.; Jones, J.R.; Frederickson, L.H.; Kaiser, M.S. Use of boreal forested wetlands
by Pacific loons (Gavia pacifica Lawrence) and homed grebes (Podiceps auritus L.):
relations with limnological characteristics. Hydrobiologia 279/280:171-183; 1994.
Hoyer, M.Y.; Canfield, D.E., Jr. Bird abundance and species richness of Florida lakes;
influence of trophic status, lake morphology, and aquatic macrophytes. Hydrobiologia
279/280: 107-119; 1994.
Hurlbert, S.H.; Chang, c.c.Y. Ornitholimnology: effects of grazing by the Andean flamingo (Phoenicoparrus andinus). Proc. Natl. Acad. Sci. U.S.A. 80:4766--4769; 1983.
Jeppesen, E.; Jensen, J.P.; S0ndergaard, M.; Lauridsen, T.L.; Pedersen, L.J.; Jensen, L. Top
down regulation in freshwater lakes with special emphasis on the role of fish, submerged
macrophytes and water depth. Hydrobiologia 342/343: 151-164; 1997.
Jupp, B.P.; Spence, D.H.N. Limitations of macrophytes in a eutrophic lake, Loch Leven. 2.
Wave action, sediments and waterfowl grazing. J. Ecol. 65:431--446; 1977.
Kendeigh, S.c. Energy requirements for existence in relation to size of bird. Condor
72:60-65; 1970.
Kerekes, J.1. Possible correlation of summer common loon (Gavia immer) population with
the trophic state of a water body. Verh. Int. Verein. Limnol. 24:349-353; 1990.
Kerekes, J.J. Problems associated with prediction of aquatic bird biomass from total
phosphorus concentration. Verh. Int. Verein. Limnol. 26; (in press).
Kerekes, J.J.; Pollard, J.B., eds. Aquatic birds in the trophic web oflakes. Proceedings of a
symposium held in Sackville, New Brunswick, Canada, in August 1991. Hydrobiologia
279/280; 1994.
King, J.R. Seasonal allocation of time and energy resources in birds. In: Paynter, R.A., ed.
Avian energetics. Cambridge, MA: Nuttal Ornithol. Club; 1974:4--70.
Ki0rboe, T. Distribution and production of submerged macrophytes in Tipper Grund (Ringk0bing Fjord, Denmark), and the impact of waterfowl grazing. J. Appl. Ecol. 17:675687; 1980.
Lauridsen, T.L.; Jeppesen, E.; Andersen, EY. Colonization of submerged macrophytes in
shallow fish manipulated Lake Vaeng: impact of sediment composition and waterfowl
grazing. Aquat. Bot. 46:1-15,1993.
Lauridsen, T.L.; Jeppesen, E.; S0ndergaard, M. Colonisation and succession of submerged
macrophytes in shallow Lake Vreng during the first five years following fish manipulation. Hydrobiologia 2751276:233-242; 1994.
Leah, R.T.; Moss, B.; Forrest, D.E. Experiments with large enclosures in a fertile, shallow,
brackish lake. Int. Rev. Ges. Hydrobiol. 63:291-310; 1978.
9. Grazing Birds-Macrophytes Interaction
195
Lillie, RA.; Evrard, J.0. Influence of macroinvertebrates on waterfowl utilization of wetlands
in the Prairie Pothole Region of northwestern Wisconsin. Hydrobiologia 279/280:235246; 1994.
Lodge, D.M. Herbivory on freshwater macrophytes. Aquat. Bot. 41: 195-224; 1991.
McKinnon, S.L.; Mitchell, S.P. Eutrophication and black swan (Cygnus atratus Latham)
populations: tests of two simple relationships. Hydrobiologia 279/280: 163-170; 1994.
McKinnon, S.L.e. The interrelationship between phytoplankton, submerged macrophytes
and black swans (Cygnus atratus) in New Zealand lakes-tests of two models. Unpubl.
MSc thesis, University of Otago, New Zealand; 1989.
Manny, B.A.; Johnson, W.C.; Wetzel, RG. Nutrient additions by waterfowl to lakes and
reservoirs: predicting their effects on productivity and water quality. Hydrobiologia
279/280:121-132; 1994.
Marion, L.; Clergeau, P.; Brient, L.; Bertru, G. The importance of avian-contributed nitrogen (N)
and phosphorus (P) to Lake Grand-Lieu, France. Hydrobiologia 279/280: 133-147; 1994.
Marriott, RW.; Forbes, D.K. The digestion oflucerne chaff by Cape Barren geese, Cereopsis novaehollGluliae Latham. Aust. 1. Zool. 18:257-263; 1970.
Mathiasson, S. A moulting population of non-breeding mute swans with special reference to
flight feather moult, feeding ecology and habitat selection. Wildfowl 24:43-53; 1973.
Mattocks, J.e. Goose feeding and cellulose digestion. Wildfowl 22: 107-113; 1971.
Mitchell, S.P. Primary production in a shallow eutrophic lake dominated alternatively by
phytoplankton and by submerged macrophytes. Aquat. Bot. 33: 101-110; 1989.
Mitchell, S.P.; Wass, R.T. Food consumption and faecal deposition of plant nutrients by
black swans (Cygnus atratus Latham) in a shallow New Zealand lake. Hydrobiologia
306: 189-197; 1995.
Mitchell, S.P.; Wass, R.T. Grazing by black swans, physical factors and the growth and loss
of aquatic vegetation in a shallow lake. Aquat. Bot. 55:205-215; 1996a.
Mitchell, S.P.; Wass, RT. Quantifying herbivory-grazing consumption and interaction
strength. Oikos 76:573-576; 1996b.
Mitchell, S.P.; Hamilton, D.P.; MacGibbon, W.S.; Nayar, P.K.B.; Reynolds, RN. Interrelations
between phytoplankton, submerged macrophytes, black swans (Cygnus atratus) and zooplankton in a shallow New Zealand lake. Int. Rev. Ges. Hydrobiol. 73: 145-170; 1988.
Moss, B.; Leah, R.T. Changes in the ecosystem of a guanotrophic and brackish shallow lake
in eastern England; potential problems in its restoration. Int. Rev. Ges. Hydrobiol.
67:625-659; 1982.
Murphy, S.M.; Kessel, B.; Vining, L.J. Waterfowl populations and limnologic characteristics of taiga ponds. J. Wild!. Manage. 48: 1156-1163; 1984.
Nagy, K.A. Field metabolic rate and food requirement scaling in mammals and birds. Ecol.
Monogr. 57:111-128; 1987.
National Environmental Research Institute (Denmark). Report and activities 1993-94.
Roskilde, Denmark: Ministry ofthe Environmen'~ 1994.
National Research Council (USA). Restoration of Aquatic Ecosystems-Science, Technology and Public Policy. Washington, DC: National Academy Press, USA; 1992.
Nilsson, L. Breeding waterfowl in eutrophicated lakes in south Sweden. Wildfowl 29:101110; 1978.
Nilsson, S.G.; Nilsson, LN. Breeding bird community densities and species richness in
lakes. Oikos 31:214-221; 1978.
Noyes, J.H.; Jarvis, RL. Diet and nutrition of breeding female redhead and Canvasback
ducks in Nevada. J. Wildl. Manage. 49:203-211; 1985.
Organisation for Economic Cooperation and Development (OECD). Eutrophication of
waters-monitoring, assessment and control. Paris: OECD; 1982.
Owen, M.; Black, J.M. Waterfowl biology. Glasgow: B1ackie; 1990.
Perrow, M.R.; Schutten, J.; Howes, J.R; Holzer, T.; Madgwick, P.J. Interactions between
coot (Fulica atra) and submerged macrophytes: the role of birds in the restoration
process. Hydrobiologia 342/343 :241-256; 1997.
196
S.E Mitchell and M.R. Perrow
Peters, R.H. The ecological implications of body size. Cambridge Studies in Ecology, 2.
Cambridge: Cambridge University Press; 1983.
Poysa, H. Resource utilization pattern and guild structure in a waterfowl community. Oikos
40:295-307; 1983.
Rees, E.C. Bewick's swans: their feeding ecology and coexistence with other grazing
Anatidae. J. App!. Ecol. 27:939-951; 1990.
Roijackers, R.M.M. Phytoplankton studies in a nymphaeid-dominated system: with special
reference to the effects of the presence of nymphaeids on the functioning and structure
of the phytoplankton communitiesl Rudolphus Martinus Maria Roijackers.-(S.I. : s.n.)
(Meppel : Krips Repro).-Ill. Thesis Nijmegen. With ref.-With summary in Dutch;
1985.
Scheffer, M.; Hosper, S.H.; Meijer, M-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Eco!. Evo!. 8:275-279; 1993.
Schutten, J.; van der Velden, J.A.; Smit, H. Submerged macrophytes in the recently
freshened lake system Volkerak-Zoom (The Netherlands), 1987-1991. Hydrobiologia
275/276:207-218; 1994.
Sedinger, J.S.; White, J.G.; Manns, EE.; Burris, EA.; Kedowski, R.A. Apparent metabo!ising of alfalfa components by herbivorous yearling pack brant. J. Wild!. Manage.
53:726-734; 1989.
Staicer, c.A.; Freedman, B.; Srivastava, D.; Dowd, N.; Kilgar, J.; Hayden, J.; Payne, E;
Pollock, T. Use of lakes by black duck broods in relation to biological, chemical, and
physical features. Hydrobiologia 2791280: 185-199; 1994.
Suter, W. Overwintering waterfowl on Swiss lakes: how are abundance and species richness
influenced by trophic status and lake morphology? Hydrobiologia 2791280: 1-14; 1994.
S!')ndergaard, M.; Kristensen, P.; Jeppesen, E. Phosphorus release from suspended sediment
in the shallow and wind-exposed Lake Arres!,), Denmark. Hydrobiologia 228:91-99;
1992.
S!')ndergaard, M.; Bruun, L.; Lauridsen, T.; Jeppesen, E.; Madsen, T.V. The impact of
grazing waterfowl on submerged macrophytes: in situ experiments in a shallow lake.
Aquat. Bot. 53:73-84; 1996.
van Donk, E.; Otte, A. Effects of grazing by fish and waterfowl on the biomass and species
composition of submerged macrophytes. Hydrobiologia 340:285-290; 1996.
van Donk, E.; De Deckere, E.; Klein Breteler, J.G.P.; Meulemans, J.T. Herbivory by
waterfowl and fish on macrophytes in a biomanipulated lake: effects on long term
recovery. Verh. Int. Verein. Lirnnol. 25:2139-2143; 1994.
van Wijk, R.J. Ecological studies on Potamogeton pectinatus L. 1. General characteristics,
biomass production and life cycles under field conditions. Aquat. Bot. 31 :211-258;
1988.
Verhoeven, J.T.A. The ecology of Ruppia-dominated communities in western Europe. III.
Aspects of production, consumption and decomposition. Aquat. Bot. 8:209-253; 1980.
von Bertalanffy, L. General system theory. New York: Braziller; 1968.
Wallace, J.B.; O'Hop, J. Life on a fast pad: waterlily leaf beetle impact on waterlilies.
Ecology 66:1534-1544; 1985.
Williams, M.J. The status and management of black swans Cygnus atratus Latham at Lake
Ellesmere since the "Wahine" storm, Apri11968. N.Z. J. Ecol. 2:34-41; 1979.
Woollhead, J. Birds in the trophic web of Lake Esrom, Denmark. Hydrobiologia 2791
280:29-38; 1994.
Zar, J.H. The use of the allometric model for avian standard metabolism-body weight
relationships. Compo Biochem. Physiol. 29:227-234; 1969.
10. Effects of Submerged Aquatic Macrophytes on
Nutrient Dynamics, Sedimentation, and Resuspension
John W. Barko and William F. James
Introduction
Accelerated eutrophication due to excessive nutrient (particularly P) loadings has
led to great interest in the role of submerged macrophytes in the nutritional
economy of freshwater aquatic systems. Submerged macrophytes are unique
among rooted aquatic vegetation because they link the sediment with overlying
water. This linkage is responsible for great complexities in nutrition and has
important implications for nutrient cycling. Despite increased attention to vegetated shallow water systems within the past 20 years, no consensus exists on
whether submerged macrophytes function as sources or sinks for particular nutrients. As a result, it has been necessary to evaluate quantitatively nutrient sourcesink relationships, involving both soluble and particulate nutrient fractions.
In this chapter, we address sediment nutrient interactions with submerged
macrophyte growth and metabolism, macrophyte influences on littoral-pelagic
nutrient dynamics, and macrophyte influences on sedimentation/sediment resuspension. Our objective is to integrate pertinent information, primarily from our
own studies, but with attention to related and complementary work of others. Our
intention is not to provide an exhaustive review of the literature in these areas.
We focus primarily on shallow freshwater systems (lakes, reservoirs, and
rivers) in northern temperate regions in which rooted submerged macrophytes
constitute structurally and metabolically important components of the underwater
landscape. Furthermore, our focus is limited to physical and chemical relation197
198
J.w. Barko and W.E James
ships. We address the role of macrophytes on nutrient budgets with limited
attention to potential influences on phytoplankton. The role of macrophytes in
affecting phytoplankton is considered more comprehensively in S0ndergaard and
Moss (this volume, Chapter 6).
Sediment Nutrient Interactions with Macrophyte Growth
For many years, controversy has existed regarding the role of roots versus shoots
and sediment versus open water in the nutrition of submerged aquatic macrophytes
(reviewed by Sculthorpe, 1967; Carignan and Kalff, 1980; Denny, 1980; Smart
and Barko, 1985; Agami and Waisel, 1986; Barko et al., 1986, Barko et al., 1991).
Quantification of the relative contribution of sediment and water to nutrient uptake
by submerged macrophytes has been critical to an improved understanding of
littoral nutrient cycling and littoral-pelagic nutrient exchanges. Based on a variety
of information sources, a generalized synthesis of sources of nutrient uptake by
rooted submerged macrophytes was provided in Barko et ai. (1991), as summarized below.
Phosphorus and nitrogen have been studied most extensively, and for these
nutrients, sediment is the primary source for uptake. Sediment appears to be the
principal site for uptake of iron, manganese, micronutrients, and trace metals as
well (Barko and Smart, 1986; Jackson et aI., 1994a,b). These latter elements tend
to coprecipitate and are usually present in extremely low concentrations in oxygenated surface waters. Dissolution products of relatively abundant salts are taken
up principally from the open water. Among these elements, potassium and calcium
are potentially most important in affecting submerged macrophyte growth. Potassium can be obtained from the sediment but is taken up mainly from the open
water (Barko 1982; Huebert and Gorham, 1983; Barko et aI., 1988). Under some
conditions, potassium can be exchanged by submerged macrophyte roots for
ammonium ions in sediment (Barko et ai. 1988). Calcium is a component of the
carbonate system and plays an important role in dissolved inorganic carbon uptake
during photosynthesis (Lowenhaupt, 1956; Smart and Barko, 1986).
Given the ecological significance of N and P in aquatic systems and the
importance of sediment in supplying N and P to submerged macrophytes, it is
important to evaluate the effects of macrophyte growth on the availability of these
nutrients. Results indicate that rooted submerged macrophytes, even with relatively diminutive root systems, are capable of significantly depleting pools of Nand P
in sediments (Prentki, 1979; Trisal and Kaul, 1983; Carignan, 1985; Barko et aI.,
1988; Chen and Barko, 1988). The magnitude of nutrient reductions in sediment
due to uptake by macrophytes can be impressive. For example, Barko et al. (1988)
reported greater than 90% and greater than 30% reductions in concentrations of
exchangeable N and acid-extractable P, respectively, from sediment on which
Hydrilla verticillata was grown over two consecutive 6-week periods.
From fertilization experiments involving sediments depleted of nutrients due to
uptake by macrophytes, subsequent growth has been shown to be limited prin-
10. Nutrient Dynamics, Sedimentation, and Resuspension
199
Macrophyte
nutrients
Available
sediment
nutrients
Figure 10.1. Conceptual diagram of macrophyte influences on nutrient supply as an
interactive function of sedimentation and sediment processing.
cipally by the availability of sediment N (Anderson and Kalff, 1986; Barko et al.,
1988, 1991; Rogers et aI., 1995). However, this generalization may not apply to
oligotrophic hard water systems such as Lawrence Lake, Michigan (Moeller et aI.,
this volume, Chapter 22) or to oligotrophic systems with sandy sediments, such as
Lake Hampen, Denmark (Christiansen et aI., 1985), where P appears to be more
important in limiting macrophyte growth.
As emphasized in Barko et a1. (1991) and here (see below), sedimentation in
concert with mixing and mineralization (Fig. 10.1) provides a potentially important source of nutrient renewal in macrophyte beds, as nutrient losses due to
macrophyte uptake must be balanced by inputs to maintain the vigor of continued
macrophyte growth. The growth of Vallisneria americana on intrinsically infertile
sediments (i.e., coarse sand) in Lake Onalaska, Wisconsin, appears to depend
greatly on supply of N through sedimentation (Rogers et aI., 1995, and Rogers,
unpublished data). Notably, a near-complete collapse of this and other submerged
macrophyte species from Lake Onalaska occurred following the drought years of
1988 and 1989, when sedimentation was likely minimal. Submerged macrophytes
have since recovered in the lake during recent years with normal flow. Greater net
sedimentation with less erosion in gently sloped rather than sharply sloped littoral
regions may partially account for the relationship established between littoral
slope and the biomass of submerged macrophyte communities (Duarte and Kalff,
1986).
The vigor of submerged macrophyte beds is likely maintained by nominal
inputs of sediment providing a nutritional subsidy (Fig. 10.1). However, excessive
inputs of sediment can result in macrophyte declines due to burial or unfavorable
underwater irradiance. Because aquatic systems subject to high rates of sediment
loading are frequently turbid, with associated constraints on photosynthetic activity, macrophytes can be expected to grow best under conditions of intermediate
200
J.w. Barko and w.F. James
Sedimentation /'
/'
/'
/'
Figure 10.2. Conceptual diagram of interacting
roles of underwater light and sedimentation in
affecting macrophyte growth.
sediment loadings (Fig. 10.2). It is under these conditions that macrophytes are
most likely to have the greatest influence on nutrient dynamics.
Given the demonstrated capacity of submerged macrophytes to mobilize nutrients from sediments directly via root uptake followed by subsequent release
during seasonal senescence and decomposition, vegetation of the littoral zone
needs to be viewed as a potential direct source of nutrients to the water column
(Barko and Smart, 1980; Carpenter, 1980; Landers, 1982; Smith and Adams,
1986). High productivity and biomass turnover of macrophytes in fertile systems
contribute to high rates of nutrient mobilization from sediments, particularly with
rapidly growing species such as Myriophyllum spicatum (Smith and Adams,
1986). These processes considered collectively, in combination with effects of
macrophyte metabolic activity (see below), can have significant effects on lacustrine nutrient budgets. However, in less fertile systems, in which nutritionally
more conservative macrophyte species tend to dominate, effects on nutrient
budgets are probably less pronounced (Barko et aI., 1991). Likewise, in large
deep lakes, where macrophytes are less abundant relative to lake volume,
effects on nutrient budgets are probably negligible (Gasith and Hoyer, this
volume, Chapter 29).
Sediment Nutrient Interactions with Macrophyte Metabolism
Increased attention in recent years to the role of submerged aquatic macrophytes
in the nutritional economy (particularly P) of lacustrine systems reflects the
unparalleled importance of P in the eutrophication process (Schindler, 1974,
1977). In addition to P mobilization directly, as discussed above, submerged
macrophytes can also mobilize sediment P indirectly via metabolic activities that
alter pH and redox conditions in the surrounding water (Andersen, 1975; Drake
and Heaney, 1987; James and Barko, 1991b; James, et al., 1996).
Until recently, littoral sediments have been regarded as a net sink for P because
surface sediment layers in littoral regions are usually oxidized. Classic "iron-phosphorus" theories (Mortimer, 1941) indicate that iron oxide-hydroxides (Fe+ 3) contained in the surface micro zone of littoral sediments should have a high binding
affinity for P (Lijklema, 1977), greatly reducing the potential for P flux into the
water column. In addition, the sorption capacity for P is high, whereas P release is
201
10. Nutrient Dynamics, Sedimentation, and Resuspension
Figure 10.3. Daily variations in pH
(A), dissolved oxygen (mean, minimum, and maximum value, (B), and
estimated rates of P release from sediments (C) during the summer 1994
for the inlet region of Lake Delavan,
Wisconsin. (Modified with permission from James et aI., 1996.)
11
9
7
~~~--~--~~--~--~
20
~~~~--~~----~~~,
15
10
5
o
20
~
15
0..<:'
10
et.
Q)~
_E
COl
Q) E
E~
'g
CJ)
5
o
;;. /,
,.
t: '.,;: ~.~.
~.~: ',:
.
:2
May
Jun
Jul
Aug
Sep
1994
low for sediments with high Fe:P ratios (i.e., >15 by weight), as long as the
microzone remains oxidized (Bostrom et aI., 1982; Jensen et al., 1992).
Phosphorus release from littoral sediments can be enhanced, however, at high
pH, even under oxic conditions, by replacement of P04 with OH- on iron or
aluminum oxides according to the following general equilibrium equation:
Increases in pH from about 8.0 to about 9.0 can result in at least a doubling in
the rate of P release from oxic littoral sediments (Boers, 1991). Photosynthesis
and respiration in submerged macrophytes can result in dramatic diel variations in pH and oxygen in the water column. For example, diel pH fluctuations
between 7-10 were measured in densely vegetated Hydrilla beds in the Potomac River, Washington, DC (Carter et aI., 1988). In such emiched systems,
substantially elevated pH conditions can result in significant increases in the
rate of P release from sediments (Bostrom et aI., 1982; James and Barko,
1991b; James et aI., 1996).
202
J.w. Barko and W.F. James
30
CIi
<II
ca
25
~";
20
a."i
_E
15
~ E
10
!!?-o
III Ra~1
T SE
COl
'i5
Q)
(f)
5
0
Oxic
pH 8.5
Oxic
pH 9.1
Anoxic
pH 9.1
Figure 10.4. Variations in mean (± I SE) rates of P release from littoral sediments as a function
of pH and dissolved oxygen (oxic vs. anoxic) for sediments collected in the inlet region of Lake
Delavan, Wisconsin. Rates of P release doubled under oxic conditions as pH increased by 0.6
units. Under anoxic conditions at high pH, rates of P release increased threefold, compared with
the rate measured under oxic conditions at the same pH. (Modified with permission from James
et aI., 1996.)
From studies we conducted in a littoral inlet region of Lake Delavan, Wisconsin, it is apparent that the submerged macrophyte community can promote marked
seasonal and daily fluctuations in pH and oxygen (James et al., 1996). In this inlet,
pH became greater than 10.0 in May and early June and fluctuated near 9.0 for the
remainder of the summer (Fig. I 0.3A). During the night, several periods of anoxia
occurred near the sediment surface between late June and late July, providing a
mechanism for anoxic P release from the sediments (Fig. iO.3B). Using ranges in
the rates of P release from sediments measured in the laboratory as a function of
pH and redox conditions (Fig. 10.4), James et al. (1996) estimated pronounced
variations in the rate of internal loading of P to Lake Delavan (Fig. 10.3C). Oxic P
release from sediments under conditions of elevated pH was the dominant means
of sediment P flux in the inlet region, accounting for more than 97% of the
sediment internal load over the summer compared with only 3% contributed as a
result of P release under anoxic conditions.
Macrophyte influences that enhance P mobilization from sediment can
account for a significant portion of the total mass of P loading internally in
aquatic systems. In the inlet of Lake Delavan, for instance, submerged macrophytes mobilized about 600 kg P (about 6 mg/m2/day) from the littoral sediments during the summer indirectly by altering pH and dissolved oxygen
conditions. An additional 600 kg P was mobilized from the sediments by
submerged macrophytes directly via root uptake (James et aI., 1995). Together,
P release from sediments plus macrophyte tissue P release was equivalent to
twice the external P load contributed to Lake Delavan from the watershed
(James et aI., 1996).
10. Nutrient Dynamics, Sedimentation, and Resuspension
203
Macrophyte Influences on Littoral-Pelagic Phosphorus Dynamics
Phosphorus released from littoral sediments can result in considerable P accumulation in the water column within macrophyte beds (James and Barko,
1991b). Much of this P is in a soluble form that can be transported directly into the
upper mixed layer of adjacent pelagic regions. Linkages between macrophytemediated P mobilization in the littoral zone and nutrient dynamics in the pelagic
zone are established by horizontal water exchanges between the two zones. Mechanisms that potentially govern horizontal P transport are (1) wind-driven patterns
that lead to general water circulation (Weiler, 1978), upwelling/downwelling
phenomena, and differential deepening (Imberger and Patterson, 1990) and (2)
convectively driven patterns (Monismith et al., 1990; James and Barko, 1991a,b;
James et aI., 1994). Prentki et al. (1979) suggested that water circulation patterns
created by wind shears could transport P derived from the littoral zone into the
pelagic zone of Lake Wingra, Wisconsin, at fluxes of 0.5-5.0 mg/m2/day. Measurements further suggest that water exchanges can be facilitated also by wind
shears that create upwelling/downwelling (i.e., seiche activity) and differential
deepening of the surface mixed layer (Imberger and Parker, 1985). Convective
circulation appears to be the dominant means of driving horizontal exchanges
between littoral and pelagic zones in small, wind-sheltered lakes and embayments
because it can occur in the complete absence of wind (Stefan et al., 1989; James
and Barko, 1991a,b; James et al., 1994). Even under calm conditions, the residence time of littoral regions can be on the order of 1 day or less as a result of
convective exchanges (Stefan et al., 1989).
Convective circulation is driven primarily by horizontal temperature (and density) gradients that develop between shallow littoral regions and deeper pelagic
regions as a result of differential heating and cooling along a depth gradient
(Monismith et al., 1990; James and Barko, 1991a,b; James et aI., 1994). On a daily
basis, shallow regions of aquatic systems typically heat and cool more rapidly than
deeper regions, resulting in the development of unstable holizontal water temperature gradients (Fig. 10.5). During differential cooling, water in shallow regions
cools more rapidly than in deeper regions, resulting in the horizontal movement of
shallow water and associated nutrients as an underflow below warmer pelagic
water (Fig. 10.6). During differential heating, the opposite pattern occurs. Shallow
regions heat more rapidly than deeper regions, resulting in horizontal movement
of shallow water and associated nutrients as a surface now over cooler water
located in the pelagic zone (Fig. 10.6). The presence of submerged macrophytes in
shallow littoral regions can contribute substantially to the development of strong
thermal gradients during the day in both the vertical and lateral planes, as foliage
near the surface converts solar irradiance to heat, further promoting differential
heating, with associated potential transport of soluble nutrients due to convection
(Fig. lO.7).
Dye studies conducted at Eau Galle Reservoir, Wisconsin, serve as a good
example of linkages between macrophyte-mediated processes that mobilize sedi-
204
J.W. Barko and W.P. James
2
G
~
1:
Q)
'5
~
-...
0
-1
OJ
~
·2
11!
-3
::J
Q)
0..
E
Q)
-4
I-
-5
5
10
15
20
25
30
August 1989
Figure 10.5. Diel variations in horizontal temperature gradients between the surface waters
of the littoral and pelagic zone of Eau Galle Reservoir, Wisconsin, in August 1989. A
positive temperature gradient indicates that surface water temperatures in the littoral zone
are greater than temperatures in the pelagic zone. A negative temperature gradient indicates
that surface water temperatures in the littoral zone are less than temperatures in the pelagic
zone.
Differential
cooling
Differential
heating
Figure 10.6. Conceptual diagram of convective water exchanges driven by differential
cooling and differential heating.
------------'
205
10. Nutrient Dynamics, Sedimentation, and Resuspension
Pelagic zone
Littoral zone
0
24
23
--
23
22
E
22
'-"'
2
..c
c..
Q)
0
3
4
Figure 10.7. Longitudinal and vertical variations in water temperature in the littoral zone during
a period of differential heating. Submerged macrophytes promote the development of a thin
intensive surface mixed layer in the littoral zone through attenuation of solar radiation that results
in the establishment of unstable horizontal water temperature gradients in the surface waters.
These gradients give rise to horizontal water exchanges that transport dissolved constituents.
ment P in the littoral zone and exchanges with the pelagic zone (James and Barko,
1991b). Differential cooling at night usually results in the development of an
underflow that moves along the littoral slope (Fig. 10.8), transporting P accumulated in the water column as a result of P release from oxic littoral sediments at
high pH (i.e., pH about 9.1-9.4). Intrusion of P-rich littoral water into the pelagic
Distance along transect (m)
a
[
50
100
I
I
2
Figure 10.8. Upper panel: Contours of
water temperature eC) in the littoral and
pelagic zone and the position of a dye
cloud approximately 12 hours after dye
injection as a result of convective water
exchanges during a period of differential
cooling. The arrow indicates the initial
dye injection point. Lower panel: Contours of total phosphorus concentration
(Jlg/L) in the littoral and pelagic zones.
(Modified with permission from James
and Barko, 199Ib.)
\-------"1
4
g
.c.
15.
Q)
o
4
6
206
J.W. Barko and W.E James
zone occurs as an interflow, confined usually to the top of the metalimnion, but it
can also be mixed into the epilimnion via wind-generated turbulence. Thus, littoral
contributions to the P economy of phytoplankton communities in surface waters of
adjacent pelagic regions can be important. Internal loadings from the littoral zone
of Eau Galle Reservoir into the adjacent upper mixed layer of the pelagic zone
appear to affect not only the reservoir nutrient budget (James and Barko, 1993) but
also vertically migrating phytoplankton communities (Taylor et al., 1988; James et
al., 1992). Implications of convective circulation as a mechanism of water exchange between littoral and pelagic zones are ecologically far-reaching, because
all kinds of dissolved constituents can be moved with water.
Macropbyte Influences on Sedimentation
In addition to influences on nutrient dynamics in the water column of aquatic
systems, submerged macrophytes play an important role also in mediating sedimentation dynamics in littoral regions and in shallow lakes. The presence or
absence of submerged macrophytes in shallow regions of lakes appears to be very
important in explaining different sediment accretion and composition patterns
along depth gradients. For instance, in lakes with no vegetated littoral zone,
sediment accretion is generally greatest in deeper areas, with rates declining in
shallower water to a minimum near the shoreline, due to erosional mechanisms
that result in sediment resuspension and focusing (Davis, 1968, 1973; Davis and
Brubaker, 1973; Likens and Davis, 1975; Hakanson, 1977; Evans and Rigler,
1980, 1983; Davis et aI., 1984; Hilton, 1985; Hilton et al., 1986). However, in
lakes with vegetated littoral regions anomalies frequently occur with respect to
this generalized depositional pattern. Sediment accretion in littoral regions is
generally greater than expected (Moeller and Wetzel, 1988; Anderson, 1990), and
patterns of sediment composition are different than expected, due to macrophyte
influences on sedimentation dynamics (Petticrew and Kalff, 1991, 1992, and see
below).
In Eau Galle Reservoir, sediment accretion and composition in nonvegetated
regions vary near-linearly along a depth gradient (Fig. 10.9; James and Barko,
1990) in a manner consistent with classic theories for depositional zones (i.e.,
zones of erosion, transport, and accumulation) developed by Hakanson (1977).
U sing variations in sediment moisture content (Hakanson, 1977), three depositional zones can be identified in nonvegetated regions of Eau Galle Reservoir: (1) an
accumulation zone located between 6-9 m (moisture content >75%), (2) a transport
zone located between 3.5-6 m (moisture content, 50--75%), and (3) an erosional
zone located at nonvegetated depths between 2.5-3.5 m (moisture content <50%).
However, patterns of sediment accretion and composition in the vegetated littoral
zone at depths less than 2.5 m differ greatly from those that would be expected on
the basis of depth alone in this reservoir. High moisture content, low sediment
density, high organic matter content, and high nutrient content in the littoral
sediments parallel relatively high rates of sediment accretion compared with
207
10. Nutrient Dynamics, Sedimentation, and Resuspension
Figure 10.9. Variations in sediment accretion
rates (upper panel) and moisture content of the
surface sediment (lower panel) in Eau Galle
Reservoir. (Modified with permission from James
and Barko, 1990.)
,......
~>.
1:
C)
..I<:
......
t:
0
25
•
• Nonvegetated sites
o Vegetated sIles
20
15
10
•
•
~0
0
<{
,......
0
100
::$!.
~
<2~
80
~o o~
t:
0
60
o6)
!!!
40
00
C
~
:::2:
oo.o ~.• ••
Be 0° •
0
i·0
0
e...
....... , .
o
•••
'b\
o·
20
0
0
4
6
8
10
Depth. m
conditions in the erosional zone. These patterns collectively indicate that submerged macrophytes are influencing sedimentation dynamics in shallow regions
by reducing sediment erosion and/or promoting sediment accretion.
Aquatic macrophytes can reduce sediment resuspension and erosion and promote accretion by reducing and/or redirecting turbulent water currents (Fonseca et
aI., 1982; Gregg and Rose, 1982; Madsen and Warncke, 1983; Eckman et al.,
1989). They can also serve as effective sediment traps via interception of suspended sediment (Patterson and Brown, 1979; Wetzel, 1979; Carpenter, 1981).
Moeller and Wetzel (1988) have suggested that sedimentation of algae and periphyton from macrophyte leaf surfaces may provide an imporumt link for transfer
of nutrients absorbed from the water (by algae) to the sediment. Similarly, it has
been reported that under conditions of nutrient enrichment, decomposing filamentous algae can provide major inputs of N and P to littoral sediments (HowardWilliams, 1981). Finally, the role of aquatic macrophytes in promoting accretion
and retention of sediment and nutrients in an otherwise erosional environment
provides a positive feedback mechanism for increasing the sediment surface area
colonizable by macrophytes (Carpenter, 1981).
Macrophyte Influences on Sediment Resuspension
By inhibiting sediment resuspension and erosion, submerged aquatic macrophytes
can play an important role in regulating water quality in shallow lakes and
impoundments. These systems, in the absence of aquatic macrophytes, are often
dominated by sediment resuspension induced by wind and/or benthic fishes,
208
J.w. Barko and W.E James
400
1992
o High pool
~
.sc:
OJ
~GI
U)
300
•
Normal pool
200
• •• •
•• •
•
100
0
400
~
:i:...
.sc:
1996
•
300
OJ
~GI
U)
200
•
100
0
0
5
25
10
15
20
Wind Speed (km h-1)
30
Figure 10.10. Relationships between wind velocity and seston concentrations in Marsh
Lake (USA) during 1992 when aquatic macrophytes were absent (upper panel) and 1996
when the lake was densely vegetated with Potamogeton sp. (lower panel). The presence of
submerged macrophytes changed dramatically the critical wind threshold required to resuspend sediment in this lake at nominal pool elevations. We normalized the data with
respect to pool elevation to account for differences in sediment resuspension as a function
of water depth at our sampling location. Seston concentrations during periods when pool
elevation exceeded nominal levels, due to periods of storm inflow, are denoted by an open
circle. Seston concentrations during periods of nominal pool elevation are denoted by a
solid circle. (Modified with permission from James and Barko, 1994.)
leading to sediment-related water quality problems such as enhanced nutrient
cycling, reduced water clarity, and high phytoplankton biomass (Dillon et aI.,
1990; Maceina and Soballe, 1990; Hellstrom, 1991; Sj3ndergaard et aI., 1992). The
establishment or occurrence of macrophytes in these systems tends to be associated with a clearwater state and low phytoplankton biomass (Hosper, 1989;
Dieter, 1990; Scheffer, 1990), due to macrophyte-mediated reduction in sediment
resuspension and other macrophyte-mediated factors that inhibit phytoplankton
growth, such as provision of refugia for zooplankton, release of allelopathic
substances, and nutrient scavenging (Scheffer et al., 1993). Thus, the establishment and maintenance of aquatic macrophyte communities is critical for water
quality in these systems (Hosper and Jagtman, 1990; Hanson and Butler, 1994).
Marsh Lake, Minnesota, an impoundment on the Minnesota River, provides an
example of the role submerged macrophytes can play in reducing sediment resuspension in a shallow lake (James and Barko, 1994, and unpublished data).
Resuspension and export of sediment downstream were examined in this lake in
10. Nutrient Dynamics, Sedimentation, and Resuspension
209
1992, when submerged macrophyte biomass was absent, and in 1996, when the
entire lake was densely vegetated with Potamogeton pectinatus. Based on a
theoretical wave model (Carper and Bachmann, 1984), nearly the entire sediment
surface area (81-100%) of Marsh Lake can be disturbed by wave activity at
nominal pool elevations at wind velocities as low as 11-15 kmlhr blowing from
any direction. During the year when macrophytes were absent from the system
(1992), observed critical thresholds of wind velocity required to resuspend sediment were about 12 kmlhr at nominal pool elevations, as seston concentrations in
the water column increased substantially above this wind velocity (Fig. 10.10).
Sediment export from Marsh Lake to downstream Lac Qui Parle Reservoir was
also observed when wind velocities exceeded this critical threshold, which had a
detrimental impact by exacerbating accretion rates and accelerating reduction in
water storage capacity in this reservoir. An almost complete absence of sediment
resuspension above nominal seston levels in this lake (Fig. 10.10) was associated
with the presence of dense submerged macrophyte beds in 1996, indicating clearly
that macrophytes are beneficial in reducing the occurrence of sediment resuspension by dampening wave activity. Seston concentrations remained relatively low
at nominal pool elevations, compared to the year 1992, even at wind velocities
exceeding 20 kmIhr, which is well above the critical velocity predicted by the
wave model to cause sediment resuspension. In addition, discharge of sediment
from the lake was much less in 1996, compared with 1992.
Concluding Remarks
Nitrogen is a key element in the growth of rooted aquatic macrophytes. Thus,
attention to this particular element needs to be elevated to the same level as for
P. The role of submerged macrophytes in the N economy of aquatic systems
also needs to be more thoroughly investigated. A variety of physical, chemical,
and biological processes (e.g., sedimentation, mineralization, and particulate
matter processing by benthic invertebrates) that potentially contribute to sediment N availability need to be evaluated within the context of macrophyte
nutrition.
Rooted submerged macrophytes play an important role in nutrient cycling in
aquatic systems by mediating fluxes of nutrients from sediments into the water
column. Aquatic macrophytes mobilize nutrients directly from sediments through
root uptake and senescence. They mobilize nutrients indirectly from sediments by
causing marked fluctuations in pH and oxygen through metabolic activities, which
enhance the rate of P release from sediments. In particular, high pH values (about
9-10) associated with macrophyte photosynthesis can result in ligand exchange
with P adsorbed to iron oxide-hydroxides on sediment particles, thus enhancing
the rate of P release from sediments. During the night, productive macrophyte
beds can also deplete the water column of dissolved oxygen through respiratory
activities, thereby promoting P release from sediments under anoxic conditions.
These processes can result in substantial accumulation of P in the littoral water
210
l.W. Barko and W.p. lames
column. Water circulation induced by diel heating and cooling of surface water or
other means in aquatic plant beds facilitates nutrient exchanges with the adjacent
open water of aquatic systems. These processes can result in enhanced phytoplankton production and deterioration in water quality and must be considered in
the nutrient budgets for aquatic systems.
In shallow high-energy environments, potential negative effects on water
quality by macrophytes (i.e., enhanced nutrient cycling) may be overshadowed
by the ability of macrophytes to moderate current and wave energies, thereby
producing a positive water quality effect by reducing sediment resuspension,
turbidity, and concentrations of suspended particulate materials. Results of studies
reported here for Marsh Lake suggest that the maintenance of stands of submerged aquatic macrophytes may be an effective management tool for limiting
wind-driven sediment and associated nutrient resuspension and discharge in
shallow water systems. Thus, macrophyte growth in some lakes (particularly
shallow wind-swept basins) should perhaps be encouraged rather than discouraged, despite their often negative effects on water use.
It is apparent that submerged aquatic macrophytes, through a variety of
mechanisms, can have important influences on sedimentation dynamics and can
mediate nutrient fluxes in aquatic systems. Studies of nutrient cycling, hydraulic
transport, and sedimentation dynamics in macrophyte beds are of great value in
providing information on rates and volumes of nutrients and sediment being
exchanged with the open water of aquatic systems. Interactions between macrophytes and phytoplankton in aquatic systems need to be examined more fully
through consideration of littoral-pelagic hydraulic interactions. Hydrological
factors and watershed activities that influence seasonal dynamics and magnitudes
of sediment transport in aquatic systems need to be evaluated within the context
of their effects on submerged macrophyte growth. Finally, the effects (favorable
versus unfavorable) of aquatic plants on water quality conditions in aquatic
systems need to be considered within the context of basin morphometry, hydrology, and local climate.
Acknowledgments. We gratefully acknowledge the Danish Natural Science Research Council for sponsoring the workshop in Silkeborg, Denmark (June
1996) entitled "The Role of Submerged Macrophytes in Structuring the Biological Community and Bio-geochemical Dynamics in Lakes," at which this
work was presented. We also thank Dr. Erik Jeppesen for his assistance and
advice during manuscript preparation and Dr. John D. Madsen and two anonymous referees for reviewing the manuscript. Funding for this research was
provided by U.S. Army Engineer Waterways Experiment Station through the
Aquatic Plant Control Research Program, the Water Operations Technical
Support Program, and the Water Quality Research Program. Additional funding was provided by the U.S. Army Engineer District, St. Paul, and the Wisconsin Department of Natural Resources. Permission to publish this information
was granted by the Chief of Engineers.
10. Nutrient Dynamics, Sedimentation, and Resuspension
211
References
Agami, M.; Waisel, Y. The ecophysiology of roots of submerged vascular plants. Physiol.
Veg. 24:607-624; 1986.
Andersen, J.M. Influence of pH on release of phosphorus from lake sediments. Arch.
Hydrobiol. 76:411-419; 1975.
Anderson, M.R.; Kalff, J. Nutrient limitation of Myriophyllum spicatum growth in situ.
Freshwat. BioI. 16:735-743; 1986.
Anderson, N.J. Spatial pattern of recent sediment and diatom accumulation in a small
monomictic, eutrophic lake. J. Paleolimnol. 3:143-168; 1990.
Barko, J.W. Influence of potassium source (sediment vs. open water) and sediment composition on the growth and nutrition of a submersed freshwater macrophyte (Hydrilla
verticillata (LJ.) Royle). Aquat. Bot. 12:157-172; 1982.
Barko, J.W.; Smart, R.M. Mobilization of sediment phosphorus by submersed freshwater
macrophytes. Freshwat. BioI. 10:229-238; 1980.
Barko, J.W.; Smart, R.M. Sediment-related mechanisms of growth limitation in submersed
macrophytes. Ecology 67:1328-1340; 1986.
Barko, J.W.; Adams, M.S.; Clesceri, N.L. Environmental factors and their consideration in
the management of submersed aquatic vegetation: a review. J. Aquat. Plant Manage.
24:1-10; 1986.
Barko, J.W.; Smart, R.M.; McFarland, D.G.; Chen, R.L. Interrelationships between the
growth of Hydrilla verticillata (L.f.) Royle and sediment nutrient availability. Aquat.
Bot. 32:205-216; 1988.
Barko, J.W.; Gunnison, D.; Carpenter, S.R. Sediment interactions with submersed macrophyte growth and community dynamics. Aquat. Bot. 41:41-65; 1991.
Boers, P.C.M. The Influence of pH on phosphate release from lake sediments. Wat. Res.
25:309-311; 1991.
Bostrom, B.; Jansson, M.; Forsberg, C. Phosphorus release from lake sediments. Arch.
Hydrobiol. Beih. Ergebn. Limnol. 18:5-59; 1982.
Carignan, R. Nutrient dynamics in a littoral sediment colonized by the submersed macrophyteMyriophylium spicatum. Can. J. Fish. Aquat. Sci. 42:1303-1311; 1985.
Carignan, R.; Kalff, J. Phosphorus sources for aquatic weeds: water or sediment. Science
207:987-989; 1980.
Carpenter, S.R. Enrichment of Lake Wingra, Wisconsin, by submersed macrophyte decay.
Ecology 61:1145-1155; 1980.
Carpenter, S.R. Submersed vegetation: an internal factor in lake ecosystem succession. Am.
Nat. 118:372-383; 1981.
Carper, G.L.; Bachmann, R.W. Wind resuspension of sediments in a prairie lake. Can. J.
Fish. Aquat. Sci. 41:1763-1767; 1984.
Carter, V.; Barko, J.W.; Godshalk, G.L.; Rybicki, N.B. Effects of submersed macrophytes
on water quality in the Tidal Potomac River, Maryland. J. Freshwat. Ecol. 4:493-501;
1988.
Chen, R.L.; Barko, J.w. Effects of freshwater macrophytes on sediment chemistry. J.
Freshwat. Ecol. 4:279-289; 1988.
Christiansen, R.; Sk\:'lvmand Friis, N.J.; S\:'lndergaard, M. Leaf production and nitrogen and
phosphorus tissue content of Littorella uniflora (L.) Aschers in relation to nitrogen and
phosphorus enrichment of the sediment in oligotrophic Lake Hampen, Denmark. Aquat.
Bot. 23:1-11; 1985.
Davis, M.B. Pollen grains in lake sediments: redeposition caused by seasonal water circulation. Science 162:796-799; 1968.
Davis, M.B. Redeposition of pollen grains in lake sediment. Limnol. Oceanogr. 18:44-52;
1973.
Davis, M.B.; Brubaker, L.B. Differential sedimentation of pollen grains in lakes. Limnol.
Oceanogr. 18:635-646; 1973.
212
J.w. Barko and w.F. James
Davis, M.B.; Moeller, R.E.; Ford, J. Sediment focusing and pollen influx. In: Haworth,
E.Y.; Lund, J.W., eds. Lake sediment and environmental history. Leicester, England:
Leicester University; 1984.
Denny, P. Solute movement in submerged angiosperms. BioI. Rev. 55:65-92; 1980.
Dieter, C.D. The importance of emergent vegetation in reducing sediment resuspension in
wetlands. l Freshwat. Ecol. 5:467-473; 1990.
Dillon, P.J.; Evans, RD.; Molot, L.A Retention and resuspension of phosphorus, nitrogen,
and iron in a central Ontario lake. Can. l Fish. Aquat. Sci. 47:1269-1274; 1990.
Drake, J.e.; Heaney, S.1. Occurrence of phosphorus and its potential remobilization in the
littoral sediments of a productive English lake. Freshwat. BioI., 17:513-523; 1987.
Duarte, e.M.; Kalff, J. Littoral slope as a predictor of maximum biomass of submerged
macrophyte communities. Limnol. Oceanogr. 31: 1072-1080; 1986.
Eckman, lE.; Duggins, D.O.; Sewell, AT. Ecology of understory kelp beds. I. Effects of
kelps on flow and particle transport near the bottom. 1 Exp. Mar. BioI. Ecol. 129: 173187; 1989.
Evans, R.D.; Rigler, F.H. Measurement of whole lake sediment accumulation and phosphorus retention using lead-21O dating. Can. J. Fish. Aquat. Sci. 37:817-822; 1980.
Evans, R.D.; Rigler, RH. A test of lead-21O dating for measurement of whole lake soft
sediment accumulation. Can. J. Fish. Aquat. Sci. 40:506-515; 1983.
Fonseca, M.S.; Fisher, J.S.; Zieman, J.C.; Thayer, G.W. Influence of the sea grass, Zostera
marina L., on current flow. Est. Coast. Shelf Sci. 15:351-364; 1982.
Gregg, W.W.; Rose, F.L. The effects of aquatic macrophytes on the stream microenvironment. Aquat. Bot. 14:309-324; 1982.
Hilkanson, L. The influence of wind, fetch and water depth on the distribution of sediments
in Lake Vanern, Sweden. Can. J. Earth Sci. 14:397-412; 1977.
Hanson, M.A; Butler, M.G. Responses of plankton, turbidity, and macrophytes to biomanipulation in a shallow prairie lake. Can. J. Fish. Aquat. Sci. 51:1180-1188; 1994.
Hellstrom, T. The effect of resuspension on algal production in a shallow lake. Hydrobiologia 213:183-190; 1991.
Hilton, J. A conceptual framework for predicting the occurrence of sediment focusing and
sediment redistribution in small lakes. Limnol. Oceanogr. 30: 1131-1143; 1985.
Hilton, J.; Lishman, J.P.; Allen P.V. The dominant processes of sediment distribution and
focusing in a small, eutrophic, monomictic lake. Limnol. Oceanogr. 31:125-133; 1986.
Hosper, S.H. Biomanipulation, new perspectives for restoration of shallow, eutrophic lakes
in The Netherlands. Hydrobiol. Bull. 23:5-10; 1989.
Hosper, S.H.; Jagtman, E. Biomanipulation additional to nutrient control for restoration of
shallow lakes in The Netherlands. Hydrobiologia 200/201 :523-534; 1990.
Howard-Williams, e. Studies on the ability of a Potamogeton pectinatus community to
remove dissolved nitrogen and phosphorus compounds from lake water. J. Appl. Ecol.
18:619-637; 1981.
Huebert, D.B.; Gorham, P.R Biphasic mineral nutrition of the submersed aquatic macrophyte Potamogeton pectinatus L. Aquat. Bot. 16:269-284; 1983.
Imberger, J.; Parker, G. Mixed layer dynamics in a lake exposed to a spatially variable wind
field. Limnol. Oceanogr. 30:473-488; 1985.
Imberger, J.; Patterson, J.e. Physical limnology. Adv. Appl. Mech. 27:303-475; 1990.
Jackson, L.J.; Rowen, D.J.; Cornett, RJ.; Kalff, J. Myriophyllum spicatum pumps essential
and nonessential trace elements from sediment to epiphytes. Can. J. Fish. Aquat. Sci.
51:1769-1773; 1994a.
Jackson, L.J.; Rasmussen, lB.; Kalff, J. A mass-balance analysis of trace metals in two
weedbeds. Wat. Air Soil Pollut. 75: 107-119; 1994b.
James, W.F.; Barko, J.W. Macrophyte influences on the zonation of sediment accretion and
composition in a north-temperate reservoir. Arch. Hydrobiol. 120:129-142; 1990.
James, W.F.; Barko, J.W. Estimation of phosphorus exchange between littoral and pelagic
zones during nighttime convective circulation. Limnol. Oceanogr. 36: 179-187; 1991a.
10. Nutrient Dynamics, Sedimentation, and Resuspension
213
James, W.E; Barko, J.W. Littoral-pelagic phosphorus dynamics during nighttime convective circulation. LimnoI. Oceanogr. 36:949-960; 1991b.
James, W.E; Barko, J.W. Analysis of summer phosphorus fluxes within the pelagic zone of
Eau Galle Reservoir, Wisconsin. Lake Reserv. Manage. 8:61-71; 1993.
James, W.E; Barko, J.W. Macrophyte influences on sediment resuspension and export in a
shallow impoundment. Lake Reserv. Manage. 10:95-102; 1994.
James, W.E; Taylor, W.D.; Barko, J.W. Production and vertical migration of Ceratium
hirundinella in relation to phosphorus availability in Eau Galle Reservoir, Wisconsin.
Can. J. Fish Aquat. Sci. 49:694-700; 1992.
James, W.E; Barko, J.W.; Eakin, H.L. Convective water exchange during differential
heating and cooling: implications for dissolved constituent transport. Hydrobiologia.
294:167-176; 1994.
James, W.E; Smith, C.S.; Barko, J.W.; Field, S.J. Direct and indirect influences of aquatic
macrophyte communities on phosphorus mobilization from littoral sediments of an inlet
region in Lake Delavan, Wisconsin. Technical Report W-95-2. U.S. Army Engineer
Waterways Experiment Station, Vicksburg, MS; 1995.
James, W.E; Barko, J.W.; Field, S.J. Phosphorus mobilization from littoral sediments of an
inlet region in Lake Delavan, Wisconsin. Arch. HydrobioI. 138:245-257; 1996.
Jensen, H.S.; Kristensen, P.; Jeppesen, E.; Skytte, A. Iron:phosphorus ratio in surface
sediment as an indicator of phosphate release from aerobic sediment in shallow lakes.
Hydrobiologia 235:731-743; 1992.
Landers, D.H. Effects of naturally senescing aquatic macrophytes on nutrient chemistry and
chlorophyll a of surrounding waters. LimnoI. Oceanogr. 27:428-439; 1982.
Lijklema, L. Interaction of orthophosphate with iron (Ill) and aluminum hydroxides.
Environ. Sci. Tech. 5:537-541; 1977.
Likens, G.E.; Davis, M.B. Post-glacial history of Mirror Lake and its watershed in New
Hampshire USA: an initial report. Verh. Int. Verein. Theor. Angew. Limnol. 19:982993; 1975.
Lowenhaupt, B. The transport of calcium and other cations in submerged aquatic plants.
BioI. Rev. 31:371-395; 1956.
Maceina, M.J.; Soballe, D.M. Wind-related limnological variation in Lake Okeechobee,
FL. Lake Reserv. Manage. 6:93-100; 1990.
Madsen, T.V.; Warncke, E. Velocities of currents around and within submerged aquatic
vegetation. Arch. HydrobioI. 97:389-394; 1983.
Moeller, R.E.; Wetzel, R.G. Littoral vs profundal components of sediment accumulation:
contrasting roles as phosphorus sinks. Verh. Int. Verein. Theor. Angew. Limnol. 23:386393; 1988.
Monismith, S.; Imberger, J.; Morison, M. Convective motions in the sidearm of a small
reservoir. Limnol. Oceanogr. 35:1676-1702; 1990.
Mortimer, C.H. The exchange of dissolved substances between mud and water in lakes. J.
Ecol. 29:280-329; 1941.
Patterson, K.J.; Brown, J.M.A. Growth and elemental composition of Lagarosiphon major
in response to water and substrate nutrients. Prog. Water Techn. 2:231-246; 1979.
Petticrew, E.L.; Kalff, J. Predictions of surficial sediment composition in the littoral zone of
lakes. LimnoI. Oceanogr. 36:384-392; 1991.
Petticrew, E.L.; Kalff, J. Water flow and clay retention in submerged macrophyte beds. Can.
J. Fish. Aquat. Sci. 49:2483-2489; 1992.
Prentki, R.T. Depletion of phosphorus from sediment colonized by Myriophyllum spicatum
L. In: Breck, J.E.; Prentki, R.T.; Loucks, O.L., eds. Aquatic plants, lake management,
and ecosystem consequences of lake harvesting. Madison, WI: Institute for Environmental Studies, University of Wisconsin; 1979:161-176.
Prentki, R.T.; Adams, M.S.; Carpenter, S.R.; Gasith, A.; Smith, S.C.; Weiler, P.R. The role
of submersed weedbeds in internal loading and interception of allochthonous materials
in Lake Wingra, Wisconsin, USA. Arch. HydrobioI. Suppl. 57:221-250; 1979.
214
J.w. Barko and W.E James
Rogers, S.J.; McFarland, D.G.; Barko, J.W. Evaluation of the growth of Vallisneria americana Michx. in relation to sediment nutrient availability. Lake Reserv. Manage. 11 :5766; 1995.
Scheffer, M. Multiplicity of stable states in freshwater systems. Hydrobiologia 200/201:
475-486; 1990.
Scheffer, M.; Hosper, S.H.; Meir, M-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Eco!. Evo!. 8:275-279; 1993.
Schindler, D.W. Eutrophication and recovery in experimental lakes: implications for lake
management. Science 184:897-898; 1974.
Schindler, D.W. Evolution of phosphorus limitation in lakes. Science 195:260-262; 1977.
Sculthrope, C.D. The biology of aquatic vascular plants. London: Edward Arnold; 1967.
Smart, RM.; Barko, J.W. Laboratory culture of submersed freshwater macrophytes on
natural sediments. Aquat. Bot. 21:251-263; 1985.
Smart, RM.; Barko, J.w. Effects of water chemistry on aquatic plants: growth and photosynthesis of Myriophyllum spicatum L. Technical Report A-86-2, Environmental Laboratory, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS; 1986.
Smith, C.S.; Adams, M.S. Phosphorus transfer from sediments by Myriophyllum spicatum.
Limno!. Oceanogr. 31:1312-1321; 1986.
S!3ndergaard, M.; Kristensen, P.; Jeppesen, E. Phosphorus release from resuspended sediment in the shallow and wind-exposed Lake Arres!3, Denmark. Hydrobiologia 228:9199; 1992.
Stefan, H.G.; Horsch, G.M.; Barko, J. W. A model for the estimation of convective exchange
in the littoral region of a shallow lake during cooling. Hydrobiologia 174:225-234;
1989.
Taylor, W.D.; Barko, J.W.; James, W.E Contrasting diel patterns of vertical migration in the
dinoflagellate Ceratium hirundinella in relation to phosphorus supply in a north temperate reservoir. Can. J. Fish. Aquat. Sci. 45: 1093-1098; 1988.
Trisal, c.L.; Kaul, S. Sediment composition, mud-water interchanges and the role of
macrophytes in Dal Lake, Kashmir. Int. Rev. Ges. Hydrobio!. 68:671-682; 1983.
Weiler, P.R Littoral-pelagic exchange in Lake Wingra, Wisconsin, as determined by a
circulation model. Madison, WI: University of Wisconsin, Madison, Inst. Environ. Stud.
Rep. 100; 1978.
Wetzel, R.G. The role of the littoral zone and detritus in lake metabolism. Arch. Hydrobio!.
Beih. Ergebn. Limnol. 13:145-161; 1979.
2. Case Studies
11. Macrophyte Structure and Growth of
Bluegill (Lepomis macrochirus): Design of a
Multilake Experiment
Stephen R. Carpenter, Mark Olson, Paul Cunningham,
Sarig Gafny, Nathan Nibbelink, Tom Pellett, Christine Storlie,
Anett Trebitz, and Karen Wilson
Introduction
Experimental manipulations of whole ecosystems can be a powerful test of ecological understanding. In particular, ecosystem-scale manipulations can evaluate
basic ecological ideas in ways that complement comparative studies, models, and
smaller-scale experiments (Carpenter et ai., 1995a). From an applied perspective,
ecosystem experiments can also give unique insights into what works at a scale
directly relevant to managers (Kitchell, 1992). When management actions are
coupled with scientific studies of the response of the ecosystem, learning may lead
to improved management practices (Gunderson et al., 1995). Here we present
early results of an experiment to test the idea that nuisance macrophytes can be
managed to enhance fish growth.
In North American lakes, dense beds of an exotic macrophyte, Eurasian water
milfoil (Myriophyllum spicatum) are often associated with stunted populations of
bluegill (Lepomis macrochirus) and their predators such as largemouth bass (Micropterus salmoides) (Heck and Crowder, 1991). One explanation for this association is that dense vegetation inhibits foraging by piscivores, allowing bluegill to
reach high densities that deplete benthic invertebrates and thereby reducing
growth rates of individual bluegill (Crowder and Cooper, 1982; Trebitz et al.,
1997). Intermediate macrophyte densities may allow predators to control bluegill
densities while also providing habitat that supports good growth by surviving
bluegill (Cooper and Crowder, 1979; Wiley et ai., 1984; Trebitz and Nibbelink,
217
218
S.R. Carpenter et al.
1996; Trebitz et a!., 1997). Lake managers are interested in both controlling
milfoil and improving growth of bluegill. We proposed that both goals could be
met by harvesting channels through mi1foil beds to make bluegill more accessible
to piscivores. This hypothesis is being tested with a multilake experiment.
Our experiment uses a set of lakes in which bluegill growth rates were measured before and after manipulation. The manipulation involved harvesting channels through the macrophyte beds in some of the lakes, while leaving the other
lakes as unmanipulated controls. This experiment was scrutinized closely because
it entailed a substantial commitment of staff and resources and conspicuous
interventions in public waters. Design of the experiment was a complex process
that required us to balance scientific goals with the realities of funding and the
expectations of managers and the public. Therefore, we conducted a series of
planning exercises to clarify the scientific goals and discuss them with interested
managers and the public. These exercises consisted of ecosystem models to help
plan the program, statistical power analysis to determine the number of lakes to be
manipulated, and comparative studies to select the experimental lakes from a large
number of candidates. Here we summarize the design process, the macrophyte
manipulation, and responses of bluegill growth in the first year after manipulation.
The Role of Modeling
We used models to reach a common understanding of the problem and our
approach, to plan the macrophyte harvesting, and to synthesize results.
Early in the project, a series of meetings of managers and scientists developed
a common understanding of macrophyte-fish interactions. Our goal was to develop a conceptual framework that would help us decide what to measure and how
to manipulate the lakes. The resulting model used a daily time step to simulate
dynamics of zooplankton, four classes of benthic and plant-associated invertebrates, and size-structured populations of bluegill and largemouth bass for
10 years (Trebitz et aI., 1997). Extensive analyses of this model suggested that
harvesting a moderate amount of milfoil in narrow channels could increase growth
rates of bluegill and population densities of largemouth bass (Trebitz et a!., 1997).
A valuable result of the modeling exercise was discussion among our diverse team
of the key state variables, the time scales of ecosystem response, and a general
strategy for the experiment.
To plan the macrophyte harvests, we needed a simpler model that could be
explained to diverse stakeholders and tied closely to field data. Our approach
combined principles of fish bioenergetics with the geometry of the littoral zone
(Trebitz and Nibbelink, 1996). This model indicated that the width of cut channels
and the percentage of vegetation removed interacted to determine bluegill growth
response (Fig. 11.1). Bluegill growth is predicted to be maximized by cutting
channels 2 or 3 m wide (near the minimum width possible with available equipment) that collectively remove about one-third of the area of the milfoil bed.
These calculations are contingent on assumptions about the relative availability of
11. Macrophyte Structure and Bluegill Growth
Figure 11.1. Predicted feeding rate of
2-year-old bluegill in Fish Lake, Wisconsin, as a function of width of harvested channels and percentage of the
macrophyte bed that is removed. Feeding rates are expressed as proportion of
the maximum (Hewett and Johnson,
1992). The heavy line marks the average
feeding rate in Fish Lake before manipulation (0.45). Effect size refers to the
relative availability of invertebrate prey
in the edge habitat (which extends 2 m
from open water) versus the milfoil bed
interior (Trebitz and Nibbelink, 1996).
2
219
m edge, effect size = 2.0
8
6
4
10 20 30 40 50 60 70
% Vegetation Removed
macroinvertebrates in edge habitats versus the interior of the milfoil bed. This
variable is difficult (perhaps impossible) to measure in the field. Trebitz and
Nibbelink (1996) used many simulations over a wide range of conditions to show
that harvesting recommendations were not sensitive to assumptions about availability of invertebrate prey. Over a wide range of assumptions, optimal bluegill
growth occurred when relatively narrow channels were cut to remove 20-40% of
the area of the milfoil beds. Although these differences in growth rate may appear
small, over the lifetime of a fish they lead to large differences in size at age
(Trebitz and Nibbelink, 1996).
Several models are being used to synthesize results of the study. A well-known fish
bioenergetics model (Hewett and Johnson, 1992) will be used with growth and diet
data to test the hypothesis that consumption of bluegill by bass increased after
manipulation. An individual-based model for bluegill has been developed to assess the
implications of changes in habitat usage by bluegill after manipulation (Nibbelink,
1996).
Power to Detect Effects
Statistical power, the probability of detecting a given effect, is an important
consideration for fisheries management experiments (McAllister and Peterman,
1992). Power is defined as the probability of detecting a specified effect by
rejecting a null hypothesis of no effect at a specified significance level, given the
error variance and degrees offreedom (Cohen, 1988). Although Bayesian statistics
can evaluate our experiment more logically, usefully, and completely than frequentist statistics (Howson and Urbach, 1989), and power is a frequentist concept,
we viewed power analysis as a useful exercise. Power analysis provided us with
"rules of thumb" for choosing sample sizes and deciding which variates to include
in the study. Also, power analysis provided insurance against possible criticisms of
the experiment from a frequentist perspective.
220
S.R. Carpenter et al.
1.0
0.8
....
Q) 0.6
3:
a
0...
E3 =Growth.r~~~~t75mm.
0.4
C = Population estimate..
0.2
01
2
Effect Size
3
4
Figure 11.2. Power (the probability of detecting the effect at the 5% significance level)
versus effect size (postmanipulation value divided by the premanipulation value). These
power curves assume five manipulated lakes and five control lakes. Power curves are
shown for bluegill growth (mm) during year I (A), growth rate (g/yr) for bluegills 75 mm
long at the start of the year (B), and population density (bluegills/ha) (C).
Power analysis showed that an experiment with about 10 lakes (five treatment,
five control) was adequate to detect effects deemed important by lake managers
(Carpenter et aI., 1995b). Even smaller effects could be detected with larger
sample sizes, but resolution of such subtle effects was not thought to be worth the
cost of additional replicates. More complex experiments, including a gradient of
harvest intensities and designs to test various interactions, were considered but
rejected as too costly at this stage of the research (Carpenter et al., 1995b).
Power to detect fish growth responses was generally greater than power to
detect changes in fish population size (Fig. 11.2), because errors in measuring
growth are smaller than errors in estimating population size. Population data are
far more costly to collect than growth data. Adding a lake to the study for
population estimates cost more than 50x more person-hours than adding a lake to
the study for growth rate estimates (Carpenter et al., 1995b). This important
insight prompted us to study growth and size structure in a larger number of lakes
and to abandon population size as a response variate.
Lake Selection
The process of selecting lakes for the experiment was based on a sequence of
independent criteria. To receive initial consideration, the fish community had to be
dominated by largemouth bass and bluegill. In addition to being two of the most
important sport fish in Wisconsin, our predicted responses to macrophyte harvesting were based on expected changes in the interaction between these two species.
11. Macrophyte Structure and Bluegill Growth
221
• Experimental Lakes
E3 Nonexperimental Lakes
6
>.
U
c::
~ 4
em
LL
2
12
16
20
24
28
32
36
40
44
Mean Bluegill Growth Rate (mm/yr)
Figure 11.3. Frequency distribution of mean bluegill growth rates from 33 of the lakes
surveyed in 1993. Mean growth rates were estimated from total length versus age regressions (mean sample size ± 1 standard deviation was 60.5 ± 24.3, and mean ?- was 0.90 ±
0.04). Of the lakes surveyed, 11 were chosen for the experiment.
We sought relatively small lakes «150 ha) to facilitate sampling and manipulation. We also chose lakes that lacked surface water connections to other lakes to
eliminate the possibility that our results could be confounded by interlake fish
movements.
Drawing on the expertise of regional fish biologists and lake managers, we
generated a pool of more than 400 lakes (from a total of 15,000 lakes in Wisconsin). To this set of lakes, we applied several scientific, political, and logistical
criteria. Because we planned a manipulation of littoral zone macrophytes, we
searched for lakes that had extensive littoral zones dominated by Eurasian milfoil.
We also eliminated shallow lakes that possessed a high risk of winterkill. Winterkill events can dramatically change fish communities (Tonn and Paszkowski,
1986; Hall and Ehlinger, 1989) and could potentially confound our experimental
manipulation. For continued eligibility, a lake had to have public access and
adequate boat launch facilities. This criterion reflects agency goals of improving
public fisheries. Furthermore, we avoided lakes where our experiment might
conflict with other research and/or stocking projects.
Using these criteria, the set of 400 lakes was narrowed to 37 candidates. These
lakes were sampled in 1993 to better characterize the fish and macrophyte communities and to quantify patterns of bluegill growth. Bluegill were collected by
electroshocking, and scales were collected to back-calculate length-at-age using
the Fraser-Lee method (Tesch, 1968). Mean annual growth rates were estimated
by determining the slope of total length versus age regressions for each lake. These
222
S.R. Carpenter et al.
estimates were used to make our final selections. To maximize our ability to detect
growth responses, we chose our experimental lakes from a narrow range of
pretreatment bluegill growth rates (Fig. 11.3). We also selected lakes with slowgrowing bluegill to increase the potential for increases in growth after manipulation (Fig. 11.3). After the lakes were selected, five were chosen to be treatment
lakes and six served as unmanipulated controls. The allocation of lakes to treatment and control groups was not strictly random but appeared to be unbiased. We
had decided a priori that Fish Lake would be harvested, because it was near
Madison and could be studied more intensively than the other lakes. Despite this
nonrandom decision, the treatment and control groups were not distinguishable in
any way that seemed capable of biasing the outcome of the experiment.
Manipulation
The experiment was initiated in August 1994 when four of the five treatment lakes
were manipulated (Gibbs Lake was manipulated in 1995). The removal of macrophytes from the littoral zone of each lake was accomplished with an aquatic plant
harvester equipped with two cutting bars. The front bar cut a 3-m swath to a
maximum depth of 2.5 m. When using this bar, cut macrophytes were collected
immediately and transported to shore for removal. The other cutting bar was
specially designed to cut to a depth of 5 m. This deep bar cut a narrower swath of
2 m, and macrophytes that floated to the surface were collected during a second
pass. Choice of cutting bars depended on the depth of macrophyte growth in a lake
and the slope of the littoral zone.
Manipulations of the treatment lakes are summarized in Table 11.1. In each
lake, macrophytes were removed in a series of channels. Depending on lake size
and type of cutting bar, the number of channels ranged from 108 to 285. Mean
channel length also varied from 44 to 123 m, due to lake morphometry and depth
Table 11.1. Summary of Macrophyte Manipulations in the Five Treatment Lakes
Lake
Area
(ha)
% of area
in littoral
zone
Dates of
manipulation
Number
of
channels
Channel
width
(m)
Mean
channel
length
(m)
Created
edge (m)
% of
littoral
cleared
Fish
102.0
54.5
285
2
123.0
70,104
18.0
26.0
65.2
8/08/948/16/94
8/14/95-
158
3
60.0
18,960
16.8
172
2
43.7
15,030
21.3
127
3
93.2
33,514
15.2
108
2
46.5
10,044
17.0
Gibbs
8122/95
Heidemann
10.5
67.2
Silver
28.3
77.7
Tuma
7.7
76.9
8/30/949/01/94
811 9/948123/94
8125/948129/94
11. Macrophyte Structure and Bluegill Growth
223
of macrophyte growth. As a consequence of variation in channel length and
number, the absolute amount of edge varied widely among lakes (Table 11.1).
However, the percentage of vegetation removed from littoral zones was similar
among lakes (mean = 17.7%, coefficient of variation = 12.8%) and close to our
target of 20%.
First Year Responses
To evaluate bluegill growth responses in the first year after our manipulation, we
electroshocked each lake in autumn 1995. Scales were collected from a size range
of bluegill to back-calculate a fish's size at the start of the growing season, which
was then subtracted from the size-at-capture to estimate growth in 1995. Mean
growth rates were calculated for age-1 through age-5 bluegill in each lake. Macrophyte densities were measured by the method of Deppe and Lathrop (1992).
Results reported here exclude Gibbs Lake, which was not manipulated until 1995
and did not have as much time as the other lakes to respond to manipulation.
Probability distributions of growth differences between cut and control lakes
were calculated for the first growing season after manipulation (Fig. 11.4). These
0.16
age 5
0.12
>-
~
:.aCIS
..CI
0.08
0
....
a..
0.04
Growth Difference, Cut - Control, mm
Figure 11.4. Probability distributions of the growth difference (mm) between cut and
control lakes for bluegill aged 1, 2, 3, 4, or 5 years. Growth increments were calculated for
January-September 1995, a period that includes the first growing season after the experimental macrophyte harvests. Distributions shown here are based on six reference lakes
and four manipulated lakes (Fish, Heidemann, Silver, and Tuma).
224
S.R. Carpenter et al.
distributions are Bayesian posterior distributions based on a noninfonnative prior
distribution (Box and Tiao, 1973). Unlike conventional t-tests, which yield the
probability of being wrong if the null hypothesis is rejected, these distributions can
be interpreted directly as the probability of a given effect (Box and Tiao, 1973).
For a specified growth difference, the height of the curve is the probability of that
growth difference. The area under the curve to the right of the growth difference is
the probability of a higher growth difference, and the area to the left is the
probability of a lower growth difference. For example, the probability that age-3
fish grew more in the harvested lakes is nearly 1 (almost the entire area under the
curve is to the right of zero). By contrast, the probability that age-5 fish grew more
in the harvested lakes is about 0.5 (about half the area below the curve is right of
zero, and half is left). If the management goal is a 5-mm growth increment, we can
see that the probability of meeting the goal is very low for bluegill aged 1 and 5,
about one-third for bluegill aged 2, about 0.5 for bluegill aged 4, and more than
90% for bluegill aged 3. These distributions can also be used to compare effect
sizes other than zero. The probability of an age-3 fish growing 10 mm is more than
five times greater than the probability of an age-2 fish growing 10 mm (at 10 mm
on the X-axis, the Yvalue for age-3 fish is much higher than the Yvalue for age-2
fish).
These early results suggest that experimental harvesting had strong positive
effects on growth of age-3 bluegill, somewhat positive and variable effects on
age-4 bluegill, no effect on growth of age-5 bluegill, mildly negative effects on
growth of age-2 bluegill, and more strongly negative effects on growth of age-l
bluegill. The variable growth responses observed in our preliminary analysis may
reflect differing patterns of habitat use among age classes. Bluegill often undergo
a habitat shift in their ontogeny, in response to size-specific changes in predation
risk (Mittelbach, 1981; Werner and Hall, 1988). Small or young bluegill typically
use the littoral zone as a refuge from predatory largemouth bass. However, because vulnerability to bass predation decreases with size, larger or older bluegill
are able to forage for zooplankton in the energetically profitable pelagic zone
(Werner and Hall, 1988). Consequently, intennediate age classes would be expected to respond most strongly to experimental manipulation of edge habitat. The
fact that only age-3 bluegill responded positively may indicate that this age class
was the only one to benefit from macrophyte harvesting. Over the course of the
growing season, these bluegill increased in size from 90- to UO-mm total length.
Within this size range, individuals often use the vegetation as a refuge between
foraging bouts in open water (Werner and Hall, 1988). Increasing the amount of
edge may have significantly increased the available foraging habitat for this age
class but not changed (or even decreased) the amount of habitat for younger age
classes.
Alternatively, age-l and age-2 bluegill may not have responded because of a
natural decrease in macrophyte density in all lakes. From 1993 to 1995, we
observed a significant decrease in overall macrophyte density in all our experimentallakes. Analysis of variance indicated a significant year effect (F1,16 =
5.74, P <.05), a nonsignificant manipulation effect (F1,16 = 2.00, P >.05), and a
11. Macrophyte Structure and Bluegill Growth
225
marginal interaction (F 1,16 ::: 4.27, P "" .05), This decrease may have created de
facto channels at a small scale, which benefited age-l and age-2 bluegill equally
in both treatment and control lakes. However, age-3 bluegill may have been too
large to benefit from the macrophyte decline but of appropliate size to benefit
from the manipulation. Future analyses of bluegill growth will directly contrast
pre- and postrnanipulation growth rates and will enable us to separate the effect of
our manipulation from natural variation in macrophyte densities.
Conclusions
The planning exercises had a substantial impact on the field experiment. Modeling
studies suggested that fish growth was a key response and that substantial changes
in edge habitat and predator-prey interactions could be achieved by removing
only about 20% of the macrophyte cover. The power analysis reinforced the idea
that it was better to focus on fish growth and include more lakes in the study than
to make costly population estimates in fewer lakes. The lake selection process
revealed the natural variability that could be seen in the experiment and helped us
select a relatively homogeneous set of lakes that could be expected to respond to
macrophyte harvest. All these exercises helped us explain the experiment to
managers, the public, and scientific referees.
After all this planning, the experiment is underway and we are beginning to see
results. As is often the case with large-scale experiments, the results are more
complex than we expected (Kitchell, 1992; Carpenter et aI., 1995a). Future ecosystem responses and more detailed analyses will improve our understanding of
the responses to the manipulations. Because the experimental design has relatively
high power, we expect to determine whether macrophyte harvesting does or does
not improve bluegill growth.
Acknowledgments. We are grateful for insightful comments by Larry Crowder,
Sebastian Diehl, Avital Gasith, and Lennart Persson. The Dane County Department of Public Works built the macrophyte harvester and conducted the harvests.
This research was funded by the Federal Aid to Sport Fish Restoration Act through
the Wisconsin Department of Natural Resources, Project P.·95-P, and the NTLLTER site.
References
Box, G.E.P.; Tiao, G.c. Bayesian inference in statistical analysis. New York: Wiley; 1973.
Carpenter, S.R.; Chisholm, S.w.; Krebs, c.J.; Schindler, D.W.; Wright, R.F. Ecosystem
experiments. Science 269:324-327; 1995a.
Carpenter, S.R.; Cunningham, P.; Gafny, S.; Munoz-del-Rio, A.; Nibbelink, N.; Olson, M.;
Pellett, T.; Storlie, c.; Trebitz, A. Responses of bluegill to habitat manipulations: power
to detect effects. North Am. J. Fish. Manage. 15:519-527; 1995b.
Cohen, J. Statistical power analysis for the behavioral sciences. Hillside, NJ: Erlbaum;
1988.
226
S.R. Carpenter et al.
Cooper, W.E.; Crowder, L.B. Patterns of predation in simple and complex environments. In:
Stroud, R.H.; Clepper, H., eds. Predator-prey systems in fisheries management. Washington, DC: Sport Fishing Institute; 1979:257-267.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegills and their prey. Ecology 63:1802-1813; 1982.
Deppe, E.R.; Lathrop, R.C. A comparison of two rake sampling techniques for sampling
aquatic macrophytes. Research Management Findings 32. Madison, WI: Department of
Natural Resources; 1992.
Gunderson, L.H.; Holling, C.S.; Light, S.S., eds. Barriers and bridges to the renewal of
ecosystems and institutions. New York: Columbia University Press; 1995.
Hall, D.J.; Ehlinger, T.J. Perturbation, planktivory and pelagic community structure: the
consequences of winterkill in a small lake. Can. J. Fish. Aquat. Sci. 46:2203-2209;
1989.
Heck, K.L.; Crowder, L.B. Habitat structure and predator-prey interactions in vegetated
aquatic systems. In: Ball, S.S.; McCoy, E.D.; Mushinsky, H.R., eds. Habitat structure:
the physical arrangement of objects in space. London: Chapman & Hall; 1991 :281-299.
Hewett, S.w.; Johnson, B.L. A generalized bioenergetics model of fish growth for microcomputers. Sea Grant Institute Publication 92-250. Madison, WI: University ofWisconsin; 1992.
Howson, C.; Urbach, P. Scientific reasoning: the Bayesian approach. LaSalle, IL: Open
Court Press; 1989.
Kitchell, J.E, ed. Food web management: a case study of Lake Mendota. New York:
Springer-Verlag; 1992.
McAllister, M.K.; Peterman, R.M. Experimental design in the management of fisheries: a
review. North Am. J. Fish. Manage. 12:1-18; 1992.
Mittelbach, G.G. Foraging efficiency and body size: a study of optimal diet and habitat use
by bluegills. Ecology 62: 1370-1386; 1981.
Nibbelink, N. Explaining growth and size structure of bluegill: inverse analysis of an
individual-based model. M.S. thesis, University of Wisconsin, Madison; 1996.
Tesch, EW. Age and growth. In: Ricker, W.E., ed. Methods for assessment of fish production in fresh waters. Oxford, England: Blackwell Scientific; 1968:93-123.
Tonn, W.M.; Paszkowski, C.A. Size-limited predation, winterkill, and the organization of
Umbra-Perea fish assemblages. Can. J. Fish. Aquat. Sci. 43:194-202; 1986.
Trebitz, A.; Carpenter, S.; Cunningham, P.; Johnson, B.; Lillie, R.; Marshall, D.; Martin, T.;
Narf, R.; Pellett, T.; Stewart, S.; Storlie, c.; Unmuth, J. A model of bluegill-largemouth
bass interactions in relation to aquatic vegetation and its management. Ecol. Model.
94:139-156; 1997.
Trebitz, A.S.; Nibbelink, N. Effect of pattern of vegetation removal on growth of bluegill: a
simple model. Can. J. Fish. Aquat. Sci. 53:1844-1851; 19%.
Werner, E.E.; Hall, D.l Ontogenetic habitat shifts in bluegill: the foraging rate-predation
risk trade-off. Ecology 69:1540-1548; 1988.
Wiley, M.J.; Gordon, R.w.; Waite, S.w.; Powless, T. The relationship between aquatic
macrophytes and sport fish production in lllinois ponds: a simple model. North Am. J.
Fish. Manage. 4: 111-119; 1984.
12. Vertical Distribution of In-Benthos in
Relation to Fish and Floating-Leaved
Macrophyte Populations
Ryszard Komij6w and Brian Moss
Introduction
Several papers have described the vertical distribution of in-benthos in fresh
waters (e.g., Kajak and Dusoge, 1971; Becket et aI., 1992; van de Bund and
Groenendijk, 1994), but the reasons for such distributions of invertebrates as
midge larvae or tubificid worms in deep bottom sediments are often obscure.
One reason may be reduced abundance in the surface sediment due to fish
predation (Hershey, 1985; Lammens et aI., 1987; Kornij6w, 1997). However, a
behavioral response by the in-benthic invertebrates to avoid fish (i.e., migration to deeper sediments) may play an important role as well but has not been
tested.
Predation by fish on in-benthos is affected by the density of vegetation (Diehl,
1988, 1992). It is possible that vegetation density also modifies the effects of fish
on the vertical distribution of in-benthos. In this chapter we experimentally tested
the hypothesis that in-benthos can adjust its vertical distribution in response to
foraging pressure by fish. We carried out the study at various densities of vegetation (nymphaeids) to test whether the effect of fish on the vertical distribution of
prey can be modified by macrophytes, providing in-benthic invertebrates with a
refuge.
227
228
R. Komij6w and B. Moss
Methods
The experiment was carried out in 18 2-m 2 enclosures, made of curtain netting
(mesh size 0.5 mm), in a dense bed of Nuphar lutea (L.) Sm. at a water depth of
about 75 cm, in the small and shallow lake Little Mere in northwest England. (For
more details of the lake and the design of the enclosures, see Kornij6w, 1997.)
Three densities of the floating leaves were created in enclosures to give 10%,50%,
and 90% coverage of the water surface, by cutting superfluous leaves at the bases
of their petioles. On June 16, 1993, six perch (mean length, 14.8 cm; SD = 0.8 cm),
collected by seining in Little Mere, were added randomly to each of three enclosures at each manipulated plant density. The experiment ended on August 15.
The fish were then caught, and their stomach contents were preserved with 4%
formaldehyde solution and examined under a dissecting microscope.
Benthic macro invertebrates were sampled by means of a perspex tube (area,
15 cm 2 ; length, 120 cm), on July 12-13 and on August 10-11. From each enclosure, six cores were taken and sliced into three depth-fractions (0-2 cm,
2-5 cm, and 5-10 cm) by using the method of van de Bund and Groenendijk
(1994). The material collected was passed through a 400-!..lm sieve to concentrate
the invertebrates.
Results
The main invertebrates burrowing in the bottom sediments were Tubificidae and
larvae of Chironomus f.l plumosus L., together constituting on average more than
90% of the total in-benthos density and biomass. Chironomus larvae were the
main chironomid prey, constituting 42% of all the midge larvae found in the guts
of perch. Tubificidae mainly consisted of Euilyodrilus hammoniensis Mich. and
Limnodrillus hoffmeisteri Clap. (ca. 30 and 70%, respectively). They were not
encountered in the diet of perch.
There was a highly significant relation between animal density and depth in the
sediment (one-way ANOVA, effect of the sediment layer on the density of Tubificidae, df = 2, F = 10.89, P = .0001; effect of the sediment layer on the density of
Chironomus larvae, df = 2, F = 7.12, P = .0016), with higher densities of animals
with depth (Figs. 12.1 and 12.2). The vertical distribution patterns of different size
classes of Chironomus differed markedly, with bigger larvae being more common
at greater depth (Kornij6w, 1997).
In July, there were no interactive effects of either plant density or perch on the
density of in-benthic animals in any sediment layer (Figs. 12.1 and 12.2;
Table 12.1). In August, the densities of Tubificidae in the separate layers (0-2,
2-5,5-10 cm) still did not depend on the density of vegetation. They were similar
in fish enclosures to those in the controls in all the layers, except for the 5-1O-cm
layer where densities were positively affected by fish (Fig. 12.1; Table 12.1).
Densities of Chironomus larvae were affected by plant density in all three sediment layers, being generally lowest at highest density of vegetation (Fig. 12.2;
Figure 12.1. Vertical distribution of Tubificidae in the bottom sediments, divided into three
layers (0-2, 2-5, and 5-10 cm) in enclosures with perch (F) and in the controls (C) at low,
medium, and high densities of Nuphar leaves in July and August. Error bars represent
standard error (n = 3).
July
August
4,000
3,500
~
E
3,000
3,500
2,500
3,000
<IJ
(ij
::J
00·2 em
"0
:~
"0
.s
~2 -5 em
2,500
2,000
2,000
1,500
1,000
500
0
i2l5 - 10 em
F
C
Low
~
F
1,500
C
Medium
~~
F
C
High
1,000
500
0
F
C
Low
F
C
M,~dium
F
C
High
Plant density
Figure 12.2. Vertical distribution of Chironomus larvae in the bottom sediments, divided
into three layers (0-2, 2~5, and 5-10 cm) in enclosures with perch (F) and in the controls
(C) at low, medium, and high densities of Nuphar leaves in July and August. Error bars
represent standard error (n = 3),
R. Komij6w and B. Moss
230
Table 12.1. Two-Way Repeated-Measures ANOVAs ofthe Effects of Nuphar /utea Density and
Feeding by Perch on the Density of Chironomus Larvae and Tubificidae in Different Sediment
Layers (0---2, 2-5, 5-10 cm)
F ratio Chironomus
0---2
Source of variation
July
Plant density
Perch
Plant density x perch
August
Plant density
Perch
Plant density x perch
*p< .05;
2-5
df
5-10
F ratio Tubificidae
0---2
2-5
5-10
(sediment depth, cm)
2
0.371°S
1.877°S
0.089°S
3.606°S
O.OOI°S
0.282°S
0.208°S
1.672°S
1.509°S
0.708°S
0.456llS
0.264°S
0.213°S
1.075°S
0.840llS
1.823°S
O.002lls
0.079 llS
2
I
2
15.550***
9.044**
15.900***
5.442**
0.462llS
2.271°S
6.766**
0.09Ys
2.733°S
0.798 llS
1.803 llS
0.690llS
I.I17 lls
0.084llS
0.648°S
0.942°S
7.104*
0.445 llS
2
** P< .01; *** P< .001; ns, not significant.
Table 12.1). Perch had a significant effect on Chironomus larvae in the surface
layer at low and medium densities of Nuphar but no effect (or even a positive
effect) at high density (Fig. 12.2; Table 12.1).
Discussion
The overall vertical distribution of in-benthos found in this study corresponded to
that found by other authors (Kajak and Dusoge, 1971; Becket et aI., 1992; van de
Bund and Groenendijk, 1994). The direct influence of vegetation on Tubificidae
was negligible. Densities of Tubificidae were similar in the surficial sediments in
the presence and absence of perch but, in the deeper layers, were about twice as
high in the presence of fish than in the controls. This suggests a behavioral
response to fish predation.
The negative effect of the increased density of Nuphar leaves on the densities
of Chironomus larvae became significant in the second month of the experiment,
which suggests that at least 1 month was needed by the larvae to adjust their
density to the changed conditions. The larvae can be classified functionally as
collector-gatherers or collector-suspension feeders. Their basic food consists of
both planktonic algae and those living on the surface of the sediments. Vegetation
itself did not influence chlorophyll. concentrations in the water (R. Komij6w, B.
Moss, and J. Measey, unpublished data). However, chlorophyll concentrations
were inversely related to plant density in the presence of perch. This could have
reduced the food supply to the benthos.
The size structure of the Chironomus larvae eaten by perch, which is considered an epi-benthic predator (Mattila, 1992), was most similar to that in the top
12. Verticle Distribution of In-Benthos
231
2-cm sediment layer (Kornijow, 1997). The abundance of chironomids in the
upper sediment layer was negatively affected by perch at low and medium
Nuphar densities. This is most easily explained as a direct effect of consumption. Larvae formerly at the surface did not seem to have moved deeper to
avoid fish predation. Otherwise, larval densities in the deeper sediment layers
should have been higher in the presence of fish than in the controls. We cannot
exclude the possibility, however, that some vertical migration occurred, but
that migration into deeper layers was compensated by some predation also at
greater depth. There is evidence that perch were able to penetrate quite deeply,
for the effect of their foraging on density of some larval classes was found to
be significant even in the deepest, 5-10 cm, sediment layer (Komijow, 1997).
The lack of effect of perch on the larvae at high vegetation density supports
the view that dense vegetation can act as a refuge (Diehl, 1988, 1992; Hershey,
1985). An equally plausible explanation may be that at high vegetation density,
fish could exploit additional food resources, epiphytic invertebrates, abundantly living on the plant surface (R. Kornijow, B. Moss, and J. Measey, unpublished data). The impact of perch on the in-benthic community could then
be softened.
This chapter reveals that the behavioral responses of Chironomus larvae and
Tubificidae to perch predation differ. Chironomus larvae did not seem to migrate
to deeper sediments, as was found for Tubificidae. In addition, the paper supports
the hypothesis that a high density of vegetation may modify the influence of perch
on the vertical distribution of in-benthic prey.
Finally, it should be stressed that the results presented here, and especially
those concerning the influence of perch on the vertical distribution of in-benthos,
should be generalized with caution and only to those fish whose feeding behavior
is similar to that of perch. It is very likely that the response of benthos might differ
from that presented in this chapter if typical in-benthic feeders such as tench or
bream had been used in the experiment.
Acknowledgments. The study was supported by the Blitish Council. The
authors thank John Measey and Sabri Kilinc for their help in field and laboratory work.
References
Becket, D.C.; Aartila, T.P.; Miller, A.c. Contrasts in the density of benthic invertebrates
between macrophyte beds and open littoral patches in Eau Galle Lake, Wisconsin. Am.
Midi. Nat. 127:77-90; 1992.
Diehl, S. Foraging efficiency of three freshwater fishes: effects of structural complexity and
light. Oikos 53:207-214; 1988.
Diehl, S. Fish predation and benthic community structure: the role of omnivory and habitat
complexity. Ecology 73:1646-1661; 1992.
Hershey, A.E. Effects of predatory sculpin on the chironomid communities in an arctic lake.
Ecology 66:1131-1138; 1985.
Kajak, Z.; Dusoge, K. The regularities of vertical distribution of benthos in bottom sediments of three Masurian lakes. Ekol. Pol. 19:485-499; 1971.
232
R. Komij6w and B. Moss
Komij6w, R. The impact of predation by perch on the size-structure of Chironomus larvae:
the vertical distribution of the prey, and habitat complexity. Hydrobiol. 342/343:207213; 1997.
Lammens, E.H.R.R.; Geursen, J.; McGiIIavry, P.J. Diet shifts, feeding efficiency and
coexistence of bream (Abramis brama), roach (Rutilus rutilus) and white bream (BUcca
bjorkna) in eutrophicated lakes. Proceedings of the Fifth Congress of European Ichthyology, 1985, Stockholm: 153-162; 1987.
Mattila, J. Can fish regulate benthic communities on shallow soft-bottoms in the Baltic Sea?
The role of perch, ruffe and roach. PhD dissertation, Abo Akademi University, Turku,
Finland; 1992.
van de Bund, W.J.; Groenendijk, D. Seasonal dynamics and burrowing of littoral chi ronomid larvae in relation to competition and predation. Arch. Hydrobiol. 132:213-225;
1994.
13. Horizontal Migration of Zooplankton:
Predator-Mediated Use of Macrophyte Habitat
Torben L. Lauridsen, Erik Jeppesen, Martin S~ndergaard,
and David M. Lodge
Introduction
Aquatic macrophytes have multiple roles in ecosystem function (Carpenter and
Lodge, 1986) and in mediating predator-prey interactions involving fish and
macroinvertebrates (Crowder and Cooper, 1982; Savino and Stein, 1982). In
recent years, authors have suggested that macrophytes also provide a spatial
refuge from fish predation for Daphnia during daytime (Timms and Moss, 1984;
Davies, 1985), thereby contributing to a lower phytoplankton level in shallow
lakes (Scheffer et aI., 1993). However, migration into macrophyte beds contradicts
earlier results suggesting that macrophytes are repellent to daphnids (Hasler and
Jones, 1949; Pennak, 1966). The evidence for the use of macrophytes as a refuge
and for the role of fish in diel horizontal migration (DHM) by zooplankton is
sparse (Davies, 1985; Yuille, 1991).
In this chapter, we describe die1 horizontal distribution of Daphnia between
structured littoral zones and open water in two lakes with fish and one without. In
addition, we report results of a laboratory experiment testing the response of
Daphnia magna to macrophytes in the presence and absence of fish and fish odor.
Study Areas and Methods
Two of the lakes (Lake Ring and Lake V<eng) are situated in central Jutland,
Denmark (see Lauridsen and Buenk, 1996). Both are shallow and eutrophic and
233
234
T.L. Lauridsen et al.
Table 13.1. Morphometric Data and Fish CPUEc
Area, ha
Mean depth, m
Max depth, m
Retention time, days
Secchi depth, m
Total P, mg PIL
CPUE c
L. Vreng
L. Ring
C. Long L. b
15
1.2
1.8
15-21
1.4
0.07
41
22
2.9
5
450
2.9
0.35
2.1
2
4
Seepage lake
3.6
0.02
No fish
10
aSecchi depth and total-P concentrations are average summer values (I May to 30 September).
Data are courtesy of S.R. Carpenter.
CCatch per unit effort, using multiple mesh size gill nets from Lake Vreng and Lake Ring.
b
contain fish (Table 13.1). The third lake, Central Long Lake, is a shallow, mesotrophic, fishless lake at the University of Notre Dame Environmental Research
Center, Gogebic County (Michigan). Morphometric and other characteristics are
given in Table 13.1.
In Lake Vreng and Central Long Lake, zooplankton were sampled once every
day (1 PM-3PM) and night (1 AM-3AM) at five stations in a 72-hour period. At each
station, sampling was undertaken in a structured environment (0.5-1 m from the
shore) and in open water (5-10 m apart from the shore) without structure. Structure consisted of tree roots (primarily Alnus glutinosa L.) and emergent macrophytes (Phragmites australis L. and Carex rostrata L.) in Lake Vreng and shrub
roots (primarily Chamaedaphne) and water lilies (Nuphar) in Central Long Lake.
Using a core sampler, we took depth-integrated samples from the surface to
3-5 em above the bottom. We sampled with a core sampler. Differences in daynight densities for each 24-hour period were recorded by means of the MannWhitney u-test. For further details, see Lauridsen and Buenk (1996).
In Lake Ring, zooplankton were sampled at 24 stations along a 45-m-Iong
transect, running 15 m from and parallel to the shore. This transect passed through
three macrophyte beds and four macrophyte-free areas. Fifteen of the stations
were located in the macrophyte beds or at the edges, and the remaining nine
stations were situated in the macrophyte-free zones. Sampling was undertaken
once every day and night as above during a 60-hour period. In the macrophytes
and the open zones, we tested whether day and night densities were different by
using the Mann-Whitney u-test. For further details, see Lauridsen and Buenk
(1996).
To test the behavioral response of Daphnia magna Straus to combinations of
the presence of macrophytes, green sunfish (Lepomis cyanellus), and sunfish odor,
we did experiments in half-filled 38-L tanks. The central oval zone of the tanks
was without macrophytes in all treatments. Daphnia, fish, or fish odor could be
added to this open zone. If fish were added, they were kept in a cage that Daphnia
were too large to enter. In treatments with macrophytes, the peripheral part of each
13. Horizontal Migration of Zooplankton
235
tank was filled with Myriophyllum excalbescens L. In all treatments, the number
of daphnids in the central open zone was counted. We tested for differences
between treatments using a Tukey's test. For further details, see Lauridsen and
Lodge (1996).
Results and Discussion
In the structured littoral in Lake V reng, two of three 24-hour periods showed
significantly higher Daphnia hyalinalgaleata densities during day than night. In
the open water, we found the reverse pattern, with significantly higher densities at
night (Fig. 13.1). At all times, Daphnia densities in the structure were higher than
in open water. These data suggest Daphnia hyalinalgaleata night-time migration
toward open water from a narrow structure-filled nearshore zone. Considered as a
whole, the narrow structure-filled zone constitutes about 0.5% of the total lake
volume. Using the maximum day-to-night differences in density, migration from
this zone may result in an increase of Daphnia density in open water of maximum
4 individualslL. Migration from open water from submerged macrophyte beds is,
however, undertaken (Lauridsen et aI., 1996), implying that limited macrophyte
coverage combined with the structured littoral zone may result in a significant
increase in Daphnia density in open water.
Results were similar in the macrophyte beds of Lake Ring: Daphnia magna
densities were significantly higher during day than night (Fig. 13.2A). In open
water, however, we did not find a reverse pattern: Daphnia density was lowest at
night (significantly in one of three 24-hour periods) (Fig. 13.2B). By testing
Daphnia densities in the macrophytes versus in open water, we found significantly
higher densities in open water at night. Sampling in the macrophyte environment
Figure 13.1. Night- and daytime mean densities (n = 5) of Daphnia hyalinalgaleata in
Lake Vreng. (A) in the structure (tree roots and
emergent macrophytes) and (B) in open water
(5 m apart from the structure). P values and
95% confidence limits given; ns, no significance. (Redrawn from Lauridsen and Buenk,
1996.)
N
o
NON
Night or day
o
T.L. Lauridsen et al.
236
c
-c
~
'-"
l
ar
E
~
,r;
~
J::.
~12
0
tU
C
~
,r;
i
0tU
0
0
N
o
o
N
Night or day
N
N
o
N
o
Night or day
N
Figure 13.2. Night- and daytime mean densities of Daphnia magna (A and B) and
Daphnia hyalina (C and D) in Lake Ring. A and C: in the structure (submerged macrophytes), n = 15. Band D: in open water, n = 9. P values and 95% confidence limits given;
ns, no significance. (Redrawn from Lauridsen and Buenk, 1996.)
about 3-5 m from the shore revealed similar results, with higher densities during
day than night (Lauridsen and Buenk, 1996), demonstrating that daphnids did not
migrate toward the shore at night. It seems likely that they migrated further out
into open water, although we do not have the data to evaluate this possibility. D.
hyalinalgaleata data from Lake Ring showed the same tendency as for D. magna,
with highest density in the macrophytes at daytime (Fig. 13.2e), although it was
only significant in one of the two 24-hour periods.
The large change in density was particularly found around and at the edges of
the macrophyte beds. From day to night the edge mean density of D. magna in
Lake Ring was reduced from 64 individualslL to 3 individualsIL, whereas in the
middle of the beds and open zones, there were no significant changes (Table 13.2).
The results demonstrate the importance of the edge as a refuge for the migrating
Daphnia species.
The weaker day-night differences for D. hyalina in Lake Ring relative to those
of D. magna are consistent with observations of Walls et al. (1990) that large
cladocerans use DHM to a greater extent. We do not have conclusive evidence for
a fish-induced DHM in the field, but we attribute the DHM of the daphnids to the
presence of planktivorous fish as there was no evidence of oxygen depletion
(Lauridsen and Buenk, 1996), which also may induce migration (Frodge et aI.,
1990). Fish-induced DHM may also explain the differentiated migration of species
of Daphnia as planktivorous fish select for large-sized prey (Eggers, 1982; Dodson, 1988).
13. Horizontal Migration of Zooplankton
237
Table 13.2. Daytime and Nighttime Mean Density of Daphnia magna at Various Stations
Inside (mac) and Outside (open) the Macrophyte Edge in Lake Ringa
Day
Night
Distance from
edge (m)
n
Mean
(nIL)
95% cl
n
Mean
(nIL)
95% cl
P
4 (mac)
I (mac)
Edge
1 (open)
3 (open)
6
12
12
12
6
15.2
47.3
63.8
45.6
28.2
15.9
26.8
39.8
24.2
8.74
6
12
12
12
6
11.8
1.5
3.0
19.7
18.5
20.4
0.7
1.9
12.3
14.4
ns
0.039
0.006
0.003
ns
aANOV A
was used to test for a time effect. As no time (block) effect was found (blockday MS = 574,
FI,46 = 2.08, P = ns; blocknight MS = 11.02, FI,46 = 0.004, P = ns), data ar,~ mean values for two
nights and two days, respectively. 95% confidence limits and P values for day-night differences are
given (Student t-test).
In previous studies, macrophytes were repellent to zooplankton (Hasler and
Jones, 1949; Pennak, 1966). Consequently, zooplankton have to choose between
fish and macrophytes. In the experiments testing behavioral response of D. magna
to the presence of macrophytes, fish, or fish odor, 15% of the Daphnia were found
in the central zone in control tanks without macrophytes. With macrophytes
present, 80% were located in the central unvegetated zone, and with fish or fish
odor present, 35% and 45%, respectively, of the daphnids were located in the
unvegetated zone (Fig. 13.3). Only the fish and the fish odor treatments were not
significantly different. These results demonstrate that avoidance of fish can increase occupancy of Myriophyllum excalbescens by D. magna and may explain why Daphnia apparently use macrophyte habitats in Lake Vreng and
Lake Ring as a refuge despite their repellent impact. The results also suggest
that the response is predominantly chemically mediated, which is consistent
with De Meester (1993).
If DHM is fish-induced, we would expect that Daphnia would avoid macrophyte night and day in fishless lakes, as they avoided macrophytes in our laboratory experiments. Consistent with this expectation, densities of Daphnia in the
fishless Central Long Lake were higher in open water than in structure in four of
Figure 13.3. Percentage of Daphnia magna in
the central open water zone in four different
treatments: I, without macrophytes; 2, with
macrophytes; 3, with fish and macrophytes; and
4, with fish odor and macrophytes. Tukey's test
results are indicated by ns between treatments
that do not differ significantly. (Redrawn from
Lauridsen and Lodge, 1996.)
2
3
Treatment
4
238
T.L. Lauridsen et al.
Figure 13.4. Night- and daytime mean densities
(n = 5) of Daphnia spp. in Central Long Lake (A)
in the structure (tree roots and emergent macrophytes) and (B) in open water (5-10 m apart from
the structure). P values and 95% confidence limits
given; ns, no significance.
A
15
o
NON
Day or night
D
N
the six samplings. However, at the nearshore stations, we found significantly
higher densities of Daphnia spp. at night in two of the three 24-hour periods
(Fig. 13.4). This pattern suggests that DHM in Central Long Lake was the reverse
of what was found in Lake Vreng and Lake Ring. The pattern also suggests that
Daphnia in Central Long Lake either were avoiding something in the open water
during night or something in the structured zone during day. We do not know what
this might be, but invertebrate predators (e.g., Chaoborus) might be a possibility.
More data on the diel horizontal patterns of Daphnia and of invertebrate predation
activity are needed to evaluate that hypothesis. Whatever the exact explanation is,
the opposite pattern of horizontal migration in Central Long Lake (relative to the
two lakes with fish) is consistent with the hypothesis that Daphnia use littoral
structure as a refuge from fish predation. However, more shallow lakes with and
without fish need to be sampled to assess the generality of the patterns that we
have discussed in this paper.
Acknowledgments. We thank Brian Moss and a second anonymous reviewer for
valuable comments and suggestions to an earlier version and Kathe M0gelvang
and Juana Jacobsen for drawings. The study was financed in part by the Centre for
Freshwater Environmental Research and the Danish Natural Science Research
Council (grant 9501315).
References
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegills and their prey. Ecology 63: 1802-1813; 1982.
Davies, J. Evidence for a diurnal horizontal migration in Daphnia hyalina lacustris Sars.
Hydrobiologia 120:103-105; 1985.
13. Horizontal Migration of Zooplankton
239
De Meester, L. Genotype, fish-mediated chemicals, and planktonic behaviour in Daphnia
magna. Ecology 74:1467-1474; 1993.
Dodson, S. The ecological role of chemical stimuli for the zooplankton: predator avoidance
behavior in Daphnia. Limnol. Oceanogr. 33:1431-1439; 1988.
Eggers, D.M. Planktivore preference by prey size. Ecology 63:381-390; 1982.
Frodge, J.D.; Thomas, G.L.; Pauley, G.B. Effects of canopy floating and submergent
aquatic macrophytes on the water quality oftwo shallow Pacific Northwest lakes. Aquat.
Bot. 38:231-248; 1990.
Hasler, A.D.; Jones, E. Demonstration of the antagonistic action of large aquatic plants on
algae and rotifers. Ecology 30:359-364; 1949.
Lauridsen, T.L.; Buenk, I. Diel changes in the horizontal distribution of zooplankton in two
shallow eutrophic lakes. Arch. Hydrobiol. 137:161-176; 1996.
Lauridsen, T.L.; Lodge, D.M. Avoidance of Daphnia magna by fish and macrophytes:
chemical cues and predator-mediated use of macrophyte habitat. Limnol. Oceanogr.
41:794-798; 1996.
Lauridsen, T.L.; Pedersen, LJ.; Jeppesen, E.; Sl'lndergaard, M. The importance of macrophyte bed size for cladoceran composition and horizontal migration in a shallow lake. J.
Plankton Res. 18:2283-2294; 1996.
Pennak, R.W. Stucture of zooplankton populations in the littoral macrophyte zone of some
Colorado lakes. Trans. Am. Microsc. Soc. 85:329-349; 1966.
Savino, J.E; Stein, R.A. Predator-prey interaction between largemouth bass and bluegills
as influenced by simulated, submersed vegetation. Trans. Am. Fish. Soc. 111 :255-266;
1982.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evol. 8:275-279; 1993.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Yuille, Th. Abundance, standing crop and production of microcrustacean populations
(Cladocera, Copepoda) in the littoral zone of Lake Biel, Switzerland. Arch. Hydrobiol.
123:165-185; 1991.
Walls, M.; Rajasilta, M.; Sarvala, J.; Salo, J. Diel changes in horizontal microdistribution of
littoral cladocera. Limnologica 20:253-258; 1990.
14.
Changing Perspectives on Food Web
Interactions in Lake Littoral Zones
Larry B. Crowder, Elizabeth W. McCollum, and
Thomas H. Martin
Introduction
New information has modified how we view food web interactions in lake littoral
habitats as well as the linkages from the littoral to pelagic habitats. First, productivity of epiphytes in lakes is extremely high, often exceeding production by
submerged macrophytes (Wetzel, 1990; Wetzel and S0ndergaard, this volume, Chapter 7). Second, a portion of this epiphyte productivity is harvested by littoral
herbivores, including snails and grazing insects (Bronmark and Vermaat, this
volume, Chapter 3; Jones et at, this volume, Chapter 4) and ultimately transported
by fish out of the littoral habitat to the pelagic food web (Schindler et al., 1996;
Vanni, 1996). Finally, food web cascades documented in the pelagia of lakes
(Carpenter and Kitchell, 1993) may also occur commonly within the littoral
habitat. Food web interactions within the littoral habitat, the importance of linkages among littoral habitat patches, and links from the littoral to the pelagic are
still poorly understood, but progress in the past 15 years has been rapid.
Species interactions (e.g., competition, predation) cannot be fully evaluated
outside a food web context because their outcomes can be modified by other
members of the web as well as variable environmental factors (Crowder et at,
1988; Winemiller and Polis, 1996). Over the past 30 years, aquatic ecologists
repeatedly have examined predation and competition by using experimental
manipulations, but the results often included unanticipated effects (Sih et at,
1985; Kerfoot and Sih, 1987). Most of these surprises relate to having not con240
14. Food Web Interactions in Lake Littoral Zones
241
sidered all the important direct and indirect effects in the food web. A food web is
"a network of consumer-resource interactions among a group of organisms, populations or aggregate trophic units" (Winemiller and Polis, 1996). Interactions in
food webs are often complex, resulting from multiple pathways linking organisms
and abiotic resources. These interactions tend to be complex because they involve
indirect effects and time lags as well as spatial and temporal variability in per
capita interaction strength (Carpenter, 1988; Paine, 1992). Identifying and measuring the strength of interactions via controlled experiments is a difficult but necessary approach to understanding food web dynamics.
The interaction of predation and resource limitation as factors determining the
structure and function of aquatic communities is not well understood. Traditionally, ecologists have considered the relative importance of resource-based (bottomup) and predator-based (top-down) forces in the structure and function of food
webs (Hunter and Price, 1992; Power, 1992; Strong, 1992). But the interaction of
these processes is more interesting and informative than some reasonable simplistic assessment of their relative importance (Carpenter et aI., 1987; Vanni, 1987,
1996). One of the first efforts to explain complex interactions between predation
and resource limitation in food webs is the "trophic cascade" hypothesis (Paine,
1980; Carpenter et al., 1985, 1987; Carpenter and Kitchell, 1993). Although the
trophic cascade has been demonstrated for relatively simple food chains in the
pelagia of lakes, the predictions sometimes fail in more complicated webs (Strong,
1992). Many factors can contribute to this failure, including omnivory, temporal,
and spatial variability; physical refuges; size structure; ontogenetic shifts among
predators and prey; and chemical defenses of plants to herbivory. Many of these
factors could be important in the submerged macrophyte habitat of lakes. Do
trophic cascades occur in the littoral habitat of lakes?
Macrophytes Provide Habitat Structure
Early on, we viewed macrophytes as adding physical complexity to lake littoral
zones and coastal marine systems (Crowder and Cooper, 1982; Heck and Crowder, 1991). Not only are macrophytes common but they are also a very productive
habitat for invertebrates; animal abundances in vegetated habitats are frequently
several orders of magnitude higher than in nearby unvegetated areas. The mechanisms most often proposed to explain the extraordinary richness of the fauna on
submerged macrophytes are (1) food supplies are greater in vegetated areas than
elsewhere (as is borne out by recent information of epiphyte productivity) and (2)
survival rates of potential prey items are greater in vegetation than elsewhere due
to the refuge provided by dense vegetation. Predator efficiency declines with
increased structural complexity (Heck and Crowder, 1991), but prey density and
diversity tend to increase with increased structure-this leads to the hypothesis
that benthic ally feeding fishes might eat more and grow better at intermediate
vegetation densities (Crowder and Cooper, 1982). Macrophyte density mediates
predator-prey interactions between bluegill sunfish (Lepomis macrochirus) and
242
L.B. Crowder, E.W. McCollum, and T.H. Martin
their invertebrate prey (Crowder and Cooper, 1982). Sparse macrophytes harbored
few prey, and dense macrophytes reduced encounter rates and pursuit and capture
probabilities. Intermediate densities of macrophytes provided the highest feeding
and growth rates to these littoral fishes (Crowder and Cooper, 1982). But our
experiment also provided several surprising results; due to decreases in the abundance of invertebrate predators (e.g., odonates) in the presence of fish, some
smaller invertebrate predators increased. Many groups that are typically prey of
blue gills declined, but others did not. It was clear even then that indirect effects
and behavioral responses of prey complicated the interpretation of our results.
Food Web Interactions in the Macrophyte Habitat
Food web interactions in vegetated littoral zones are complex. Fish can suppress
the abundance of large invertebrate predators or grazers potentially leading to
indirect (and sometimes cascading) effects (Heck and Crowder, 1991). Until
recently, few people have focused on the whole submerged macrophyte food web.
Carpenter and Lodge (1986) proposed some hypotheses on the effects of submerged macrophytes on ecosystem processes in freshwater. Heck and Crowder
(1991) extended and modified these hypotheses to deal directly with the differences between webs dominated by meso grazers and those dominated by macrograzers. A typical submerged macrophyte ecosystem is sketched in Figure 14.1.
Obviously, particular systems will diverge from this figure, and many species shift
from one role to another through ontogeny. Large predators such as fish, birds, and
some mammals consume small fish and invertebrate predators that eat small
herbivores (mesograzers). Large predators may also consume large herbivores
(macrograzers). In many systems, the herbivores include both mesograzers (primarily oriented toward epiphytic algae) and macrograzers (primarily oriented
toward macrophytes). Epiphytes live on the surface of macrophytes and may
derive some nutrients directly from them. But epiphytes and macrophytes can also
compete for nutrients (particularly those in the water column) or for light.
In a system containing only mesograzers, the hypothetical effect of decreased
abundance of large predators (e.g., due to harvesting) would be that small predators would increase, leading to reductions in mesograzers and increases in
epiphyte biomass (Figure 14.2). If the epiphyte load becomes too high, light
limitation could lead to losses of submerged vegetation. This parallels the pelagic
trophic cascade (Carpenter et a!., 1985) and is an alternative explanation to
nutrient enrichment for increase in epiphyte biomass and declines in submerged
macrophytes. This system may tend to stabilize because as macrophytes are
reduced, small predators associated with the vegetation are more exposed to
predation by large predators. This should allow some recovery of the meso grazers
and some reductions in epiphyte biomass, benefiting the macrophytes (Heck and
Crowder, 1991).
In systems dominated by macro grazers, the hypothetical effects of reductions
in large predators would be increases in macrograzers, which could lead directly
14. Food Web Interactions in Lake Littoral Zones
.......................................
243
Large Predators
(large fish, birds, mammals)
l------~ IS~;~~:r~~;~~esJ~ ------:,
•_
Mesograzers
(snails, amphipods)
1r
g;r~c!.P~d.9t;~n _
_ _ _
Macrograzers
(crayfish)
1 tI
''6
~
,~
-- -, a.
"·'1
i
:i
'8
I'" --- -- --- - IMacro~ - - i -,
"'-,
//
i
Epiphytes
nutrients
i............................................... ~ & Nutrients
r---'~:
I.·. ·. .·..... · . ·...·. . ·. · . ·. !
Figure 14.1. Generalized food web for submerged macrophyte systems. Boxes contain the
major types of interactors. Lines ending in arrows indicate positive effects to the upper
trophic level of energy flow up the web. Lines ending in closed circles indicate negative
effects on the lower trophic level due to the level above. These may include the potential for
predators to "control" prey densities. Dashed lines indicate potential interactions such a leaking
of nutrients from macrophytes to epiphytes or indirect effects of cover on predator-prey
interactions. Dotted lines indicate possible routes of nutrient recycling from consumers.
to losses of macrophytes along with their associated epiphytes and fauna
(Fig. 14.2). Reduced refuge might also lead to decreased survival of small fish,
thus leading to further reductions in the abundance of large fish through poor
recruitment. This likely explains why systems dominated by macrograzers that can
grow large enough to be relatively immune from predation will have two alternative
states: if predators are abundant, macrograzers will be low and macrophytes abundant.
Alternatively, if predators are low, macrograzers are abundant and macrophytes are
sparse (Carpenter and Lodge, 1986). Many macrograzers are omnivorous and may
have other food web effects, particularly on the abundance of mesograzers (in systems
for which both grazer types are common) (Lodge et al., 1994).
Another set of linkages involve consumer-mediated nutrient recycling (and
transport). Both predators and grazers influence prey directly via consumption, but
they also release nutrients that can affect the production and potentially the
composition of the algal community (Fig. 14.1). Predators and grazers can also
alter the ratios of nutrients available (Sterner, 1990; Sterner et al., 1996), benefiting some producers over others. The idea of consumer-mediated nutrient recycling
has received some attention in the pelagic food web of lakes (Vanni, 1996) but
basically has not been explored in vegetated systems.
244
L.B. Crowder, E.w. McCollum, and T.R. Martin
Large predators
Small predators
Mesograzer
dominated
Macrograzer
dominated
1
1
t
Macrograzers
Mesograzers
Epiphytes
Macrophytes
1
t
I
t
t
I
,1
Figure 14.2. Theoretical food web responses to reductions in the abundance of large
predators. The left column reflects the response in mesograzer-dominated webs and the
right had column reflects macrograzer-dominated webs. lResponse is dependent on the
magnitude and sign of feedback from macrograzers.
Evidence for a Littoral Trophic Cascade
We conducted a 16-month field experiment to examine direct and indirect effects
of fish on their littoral prey assemblage in Bay's Mountain Lake, Tennessee. The
experiment was conducted in 24-m2 mesh enclosures in which we manipulated the
presence and absence of large redear sunfish (Lepomis microlophus > 150 mm
standard length [SL]) and small sunfish (L. macrochirus and L. microlophus
<50 mm SL). We found that both large redear sunfish and small sunfish suppressed the recruitment of snails to experimental enclosures, but snails (primarily
Helisoma anceps, Physella heterostropha, and Gyraulus parvus) increased significantly during the first 2 months of the experiment in the fish-free controls.
Five months into the experiment, the difference in snail biomass between the
enclosures with and without fish was to-fold. By the end of the experiment,
enclosures without fish had significantly lower periphyton percentage cover
(similar trends were detected in biovolume) and significantly higher macrophyte
biomass (Fig 14.3; Martin et aI., 1992). Bronmark et al. (1992) also found that fish
enhanced periphyton biomass by removing snail grazers.
Direct and Indirect Effects of Fish and Snails on Periphyton
Although Martin et aI. (1992) interpreted this littoral cascade as due to indirect
effects of fish on periphyton, mediated through snail grazers, periphyton clearly
could be enhanced directly by nutrient release from fish. To separate these effects
and determine their relati ve importance, we manipulated the presence and absence
of redear sunfish and snails, P. heterostropha, in a replicated factorial tank experiment (McCollum et al., 1998). Experimental tanks (72 L) were placed outdoors in
245
14. Food Web Interactions in Lake Littoral Zones
Figure 14.3. Littoral trophic cascade in Bay's Mountain Lake, Tennessee. Fish (both large and small
Lepomis sp.) apparently suppressed
the recuitment of snails to experimental enclosures. Without fish,
snail biomass increased, periphyton
cover and biomass declined, and
macrophyte biomass increased.
240
160
j
80
c::
o
....,
>
0
;'0
..c::
_
.2...
.,
0..
.,U
.,
I::
...
0..
a
100
75
1
I
50
25
a
Absent
Present
Fish Occurrence
two water baths, which acted as thermal buffers. Each aquarium was divided by a
plastic tank divider (perforated with I-mm holes on 5-mm centers) so that one side
of the aquarium held about one-third of the aquarium volume and the other side
held the remaining two-thirds (Fig. 14.4).
Artificial plants were used as substrata for the periphyton (Fig. 14.4). The
plants were made from bamboo skewers (the main stem) and black polypropylene
I
I
I
I
~
Figure 14.4. Aquarium set up used in experiment to separate the effects of fish feeding and
nutrient recycling on periphyton.
246
L.B. Crowder, E.W. McCollum, and T.H. Martin
FISH!
=A
=S
o
1$
FISH;
0;
=FS
=F
'$
Figure 14.5. Schematic of experimental manipulations in McCollum et ai. (1988). The
four treatments in the factorial were algae alone (A) (i.e., no animals), snails only (S) (in the
large side of the aquarium, were removed at a specified "predation rate" and fed to fish in
"fish only" treatments), fish only (F) (in the small side of the aquarium), and fish + snails
(FS) (snails removed from large side of the aquarium and fed to fish on the small side of the
same aquarium).
ribbon cut into leaves 7.6 cm long (70 cm 2); each "plant" supported 12 leaves. The
ribbons were gently abraded with emery board to create a textured surface,
allowing for a more diverse periphyton community than would have colonized
smooth ribbon (Muntenau and Malay, 1981; Pringle, 1990). The artificial plants
were constructed to mimic submerged Polygonum densiflorum, a common macrophyte in North Carolina littoral zones.
We manipulated the presence of redear sunfish and snails in a 2 x 2 factorial
experiment (Fig. 14.5); all treatments contained a periphtyon community. The four
treatments in the factorial were algae alone (A) (i.e., no animals), snails only (S)
(in the large side of the aquarium, were removed at a specified "predation rate"
and fed to fish in "fish only" treatments), fish only (F) (in the small side of the
aquarium), and fish + snails (FS) (snails removed from large side of the aquarium
and fed to fish on the small side of the same aquarium). The experiment was
blocked by water bath; treatments were replicated twice within each block, resulting
in four replicates per treatment. An algal inoculum was created by collecting macrophytes from local lakes and rinsing the associated periphyton into deionized water.
This slurry was homogenized gently in a loosely fitting tissue grinder to create a
homogeneous mixture immediately before addition to the tanks. Periphyton were
allowed to colonize and grow for 2 weeks before the initiation of treatments.
In treatments with fish, one redear sunfish was placed in the smaller section of
the aquarium; in treatments with snails, 45 snails (360 snails/m2 ) were added to the
14. Food Web Interactions in Lake Littoral Zones
247
larger section. Thus, when fish and snails were together, they were separated by a
plastic divider. To simulate fish predation, we removed three snails from each snail
tank at 3-day intervals and fed them to the fish in the fish treatments (36 snails
removed over the course of the experiment).
During the experiment, we determined nutrient concentrations and periphyton
cell number, species composition, cell size, and biovolume every 10 days. Snail
behavior (position in the water column) was recorded on five dates, and snail
number and sizes were recorded at the end of the 5-week experiment.
Complete results of this experiment are reported elsewhere (McCollum et al.,
1988). Briefly, however, concentrations of phosphorus and nilrogen in the water
were significantly higher with fish, but this had little direct effect on total periphyton cell number or biovolume. Snails decreased periphyton cell number; they
also increased average cell size of periphyton and of green algae in the absence of
fish. Snails decreased diatom and blue-green biovolume. Fish also decreased
diatom biovolume by decreasing the average cell size of diatoms. Snails increased
the percentage of the algal community comprised of gelatinous colonies. Fish
suppressed snail grazing independent of snail mortality and also inhibited snail
reproduction. Our previous research (Martin et al., 1992) suggested that fish have
a positive indirect effect on algae by removing grazers (an interaction chain)
(Wooton, 1993). Fish apparently enhance this effect by inhibiting snail reproduction and by suppressing grazing as well (interaction modifications) (Wooton,
1993). Fish can also have an important but less obvious direct effect on algae, via
nutrient recycling, possibly by altering competitive outcomes among taxa and
growth forms (McCollum et al., 1998).
Concluding Remarks
It is clear that both pelagic and littoral habitats can support trophic cascades,
but it is also clear that the interactions in these systems are complex and that
interaction chains can be modified in a wide variety of ways. Ultimately,
however, we will have to make the additional step of linking the littoral and
pelagic food webs in lakes (Lodge et aI., 1988). Animal movements (e.g.,
onshore, offshore migrations of fish, vertical migrations of zooplankters) link
the littoral and pelagic habitats both in terms of prey consumption and translocation of nutrients (Carpenter et aI., 1992; Schindler et aI., 1996; Vanni,
1996). Epiphyte productivity is high and supports substantial secondary productivity, leading to a productive littoral food web. Consumers that feed in the
littoral habitat can transform and translocate nutrients, allowing substantial
linkages from the littoral to the pelagic.
Acknowledgments. Financial support was provided by NSF BSR-8709108 (L.B.c.)
and an NSF graduate fellowship (E.W.M.). Christer Bronmark and Steve Carpenter
provided insightful comments on a prevous draft of the manuscript
248
L.B. Crowder, E.W. McCollum, and T.H. Martin
References
Bronmark, e.; Klowsiewski, S.P.; Stein, R.A Indirect effects of predation on a freshwater
benthic food chain. Ecology 73:1662-1674; 1992.
Carpenter, S.R., ed. Complex interactions in lake communities. New York: Springer-Verlag;
1988.
Carpenter, S.R.; Kitchell, J.E The trophic cascade in lakes. Cambridge: Cambridge University Press, 1993.
Carpenter, S.R.; Lodge, D.M. Effect of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Cmpenter, S.R.; Kitchen, J.E; Hodgson, J.T. Cascading trophic interactions and lake productivity. BioScience 35:634--639; 1985.
Carpenter, S.R.; Kitchell, J.E; Hodgson, J.T.; Cochran, P.A; Elser, J.J.; Elser, M.M.; Lodge,
D.M.; Kretchmer, D.; He, X.; von Ende, C.N. Regulation of lake primary productivity
by food web structure. Ecology 68: 1863-1876; 1987.
Carpenter, S.R.; Kraft, e.E.; Wright, R.A.; He, X.; Soranno, P.A; Hodgson, J.R. Resilience
and resistance of a lake phosphorus cycle before and after food web manipulation. Am.
Nat. 140:781-798; 1992.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegills and their prey. Ecology 63: 1802-1813; 1982.
Crowder, L.B.; Drenner, R.W.; Kerfoot, w.e.; McQueen, DJ.; Mills, E.L.; Sommer, U.;
Spencer, e.N.; Vanni, MJ. Food web interactions in lakes. In: Carpenter, S.R., ed.
Complex interactions in lake communities. New York: Springer-Verlag, 1988:141-160.
Heck, KL, Jr.; Crowder, L.B. Habitat structural complexity and predator-prey interactions
in vegetated aquatic systems. In: Bell, S.S.; McCoy, E.D.; Mushinsky, H.R., eds. Habitat
structure: the physical arrangement of objects in space. London: Chapman & Hall,
1991:281-299.
Hunter, M.D.; Price, P.W. Playing chutes and ladders: heterogeneity and the relative roles of
bottom-up and top-down forces in natural communities. Ecology 73:724--732; 1992.
Kerfoot, C.; Sih, A, eds. Predation: direct and indirect impacts on aquatic communities.
Hanover, NH: University Press of New England; 1987.
Lodge, D.M.; Barko, lW.; Strayer, D.; Melack, J.M.; Mittelbach, G.G.; Howarth, R.W.;
Menge, B.; Titus, J.E. Spatial heterogeneity and habitat interactions in lake communities. In: Carpenter, S.R., ed. Complex interaction in lake communities. New York:
Springer-Verlag; 1988: 181-208.
Lodge, D.M.; Kershner, M.W.; Aloi, J.E.; Covich, A.P. Effects of an omnivorous crayfish
(Orconectes rusticus) on a freshwater littoral food web. Ecology 75:1265-1281; 1994.
Martin, T.H.; Crowder, L.B.; Dumas, e.E; Burkholder, J.M. Indirect effects of fish on
macrophytes in Bays Mountain Lake: evidence for a littoral trophic cascade. Oecologia
(Berlin) 89:476-481; 1992.
McCollum, E.W.; Crowder, L.B.; McCollum, S.A Complex interactions of fish, snails and
littoral zone periphyton. Ecology 79; 1998.
Muntenau, N.; Malay, E.J. The effect of current on the distribution of diatoms settling on
submerged glass slides. Hydrobiologia 78 :278-282; 1981.
Paine, R.T. Food webs: linkage, interaction strength, and community infrastructure. 1.
Anim. Eco!. 49:667-685; 1980.
Paine, R.T. Food web analysis through field measurements of per capita interaction
strength. Nature 355:73-75; 1992.
Power, M.E. Top-down and bottom-up forces in food webs: do plants have primacy?
Ecology 73:733-746; 1992.
Pringle, e. Nutrient spatial heterogeneity: effects on community structure, physiognomy,
and diversity of stream algae. Ecology 71 :905-920; 1990.
Schindler, D.E.; Carpenter, S.R.; Cottingham, K.L.; He, X.; Hodgson, J.R.; Kitchell, J.E;
Soranno, P.A Food web structure and littoral zone coupling to pelagic trophic cascades.
14. Food Web Interactions in Lake Littoral Zones
249
In: Polis, G.A.; Winemiller, K.O., eds. Food webs: integration of pattern and dynamics.
New York: Chapman & Hall; 1996:96-108.
Sih, A.; Crowley, P.; McPeek, M.; Petranka, J.; Strohmeier, K. Predation, competition, and
prey communities: a review of field experiments. Annu. Rev. Ecol. Syst. 16:269-305;
1985.
Sterner, R.W. The ratio of nitrogen to phosphorus resupplied by herbivores: zooplankton
and the algal competitive arena. Am. Nat. 136:209-229; 1990.
Sterner, R.W.; Elser, J.1.; Chrzanowski, T.H.; Schampel, J.H.; George, N.B. Biogeochemistry and trophic ecology: a new food web diagram. In: Polis, G.A.; Winemiller,
K.O., eds. Food webs: integration of pattern and dynamics. New York: Chapman &
Hall; 1996:72-80.
Strong, D.R. Are trophic cascades all wet? Differentiation and donor control in speciose
ecosystems. Ecology 73:747-754; 1992.
Vanni, MJ. Effects of nutrients and zooplankton size on the structure of a phytoplankton
community. Ecology 68:624-635; 1987.
Vanni, M.J. Nutrient transport and recycling by consumers in lake food webs: implications
for algal communities. In: Polis, G.A.; Winemiller, K.O., eds. Food webs: integration of
pattern and dynamics. New York: Chapman & Hall; 1996:81-95.
Wetzel, R.G. Land-water interfaces: metabolic and limnological indicators. Verh. Int.
Verein. Theor. Limnol. 24:6-24; 1990.
Winemiller, K.O.; Polis, G.A. Food webs: what do they tell us about the world? In: Polis,
G.A.; Winemiller, K.O., eds. Food webs: integration of pattern and dynamics. New
York: Chapman & Hall; 1996: 1-22.
Wooton, J.T. Indirect effects and habitat use in an intertidal community: interaction chain&
and interaction modifications. Am. Nat. 141:71-89; 1993.
15. Bacterioplankton and Carbon Thrnover in a
Dense Macropbyte Canopy
Morten Ssz>ndergaard, Jon Theil-Nielsen, Kirsten Christoffersen,
Louise Schluter, Erik Jeppesen, and Martin Ssz>ndergaard
Introduction
Studies on cascading trophic interactions in lakes have shown that planktonic food
web changes may take place to the level of protozoans (reviewed by Carpenter and
Kitchell, 1993; Riemann and Christoffersen, 1993). It is more unclear if and how
cascading might influence bacterioplankton (Jeppesen et a!., 1992; Christoffersen
et a!., 1993; Pace, 1993). From studies in oligo-mesotrophic temperate lakes, Pace
(1993) concluded "that bacteria responded to changes in phytoplankton and increases in nutrients, but not to changes in zooplankton." More generally, it was
suggested that "trophic cascades do not have immediately obvious consequences
for microbial processes in lakes" (Kitchell and Carpenter, 1993). In accordance,
Jeppesen et a!. (1992) found that a trophic cascade with high grazing by cladocerans and a four- to sixfold reduction in phytoplankton biomass only slightly
altered bacterioplankton production in two fish-manipulated shallow and eutrophic Danish lakes.
Submerged macrophytes aid food web changes by their function as refuge for
zooplankton (Timms and Moss, 1984) and are a potential substrate source for
bacteria. Jeppesen et al. (1992) suggested that a high biomass of submerged
macrophytes could explain deviations at the microbial level from their tentative
model of pelagic trophic interactions. Likewise, Pace (1993) recognized organic
carbon from allochthonous and littoral sources as a possible explanation of unclear
responses by the bacterioplankton to trophic cascades. Thus, one main reason for
250
15. Bacterioplankton and Carbon Turnover
251
the apparent uncoupling ofbacterioplankton production from the pelagic food web
in some shallow lakes could be macrophytes.
There is a lack of quantitative information on the effects of macrophytes and
littoral zone production on pelagic microbial communities in lakes. Coveney and
Wetzel (1995) found that bacterioplankton production in oligotrophic Lake Lawrence could only be sustained and maintained by an input of organic substrate
from littoral production. Furthermore, low planktonic P:R ratios (<1) in oligotrophic and mesotrophic lakes, but not in eutrophic lakes, suggest littoral and
allochthonous organic carbon to be of quantitative importance to planktonic metabolism (del Giorgio and Peters, 1994). These results support the increasing recognition of nonphytoplanktonic organic sources in lakes and decoupling of direct
metabolic links between phyto- and bacterioplankton (Findlay et aI., 1992; Hessen, 1992; Kairesalo et aI., 1992; Tranvik, 1992; Wetzel, 1992).
The purpose of the present study was to compare bacterioplankton production
and planktonic microbial communities in enclosures with and without submerged
macrophytes. The use of enclosures excluded an input of organic matter from
other sources than the enclosed communities. Carbon flow scenarios with
respect to bacterioplankton carbon demand and planktonic substrate sources
were constructed.
Materials and Methods
Large circular enclosures (5-m diameter) were placed at a depth of 60 to 70 cm in
the eutrophic and shallow Lake Stigsholm, Denmark. In three enclosure areas, the
submerged macrophyte Potamogeton pectinatus reached a relative plant colonized
volume of about 47% and almost 100% area cover. Three other enclosure areas
were kept free from macrophytes. Two days before the experiment (from July 26
to August 3, 1994) the enclosure areas were closed with a curtain made of heavy
plastic. All fish were removed by electrofishing and trapping. Planktivorous fish (0+
perch) at natural lake densities (4/m2) were added to the enclosures on August 1.
Whole water column samples were taken randomly at 25-30 points within each
enclosure every second day at noon and on two occasions at midnight. All
subsamples were pooled, and processing took place either immediately at the lake
shore or within 1 hour. Vertical profiles were measured during one day/night cycle
in two enclosures.
Chlorophyll a and primary production (the 14C method) were measured according to SchWter et al. (1997). Particulate organic carbon (POC) was measured in
triplicate on GFIF filters (Sj)ndergaard and Middelboe, 1993) and dissolved organic carbon (DOC) was measured in the filtrates with a Shimadzu TOC-5000
(Sj)ndergaard et al., 1995).
Zooplankton abundance was measured with standard microscopy technique.
Mesozooplankton clearance rates were measured in short-term uptake experiments by using radio tracer labeled natural assemblages of phytoplankton and
bacteria (Jeppesen et aI., unpublished data).
252
M. S!Ilndergaard et al.
Samples for bacterial abundance were fixed with glutaraldehyde (1.5% final
conc.), stained with DAPI (Porter and Feig, 1980), and counted by epifluorescence
microscopy (at least 500 cells). Individual bacteria cells (at least 25 cells) were
sized on enlarged micrographs, and the biovolume was calculated as rods with
hemispheres. Two approaches were used to convert biovolume to biomass. In
scenario I, we used a constant of 100 fg C/llm3 (Fagerbakke et aI., 1996; TheilNielsen and S!/lndergaard, unpublished data), and in scenario II a scaling factor
according to size was included: C = 120 X VO· n , (C is carbon/cell and V is cell
volume; Norland, 1993, based on Simon and Azam, 1989). Both of the chosen
conversion factors are in current use (Carlson and Ducklow, 1996; Simek et aI.,
1996), although they cannot be considered global standards.
Bacterial cell production was measured with the thymidine method (Fuhrman
and Azam, 1980). 3H-thymidine was added to a final and saturating concentration
of 20 nM, incubated for 30 minutes, and stopped with ice-cold TCA (5% final
conc.). Cell production was calculated with the empirical factor 2 x 1018 cells/mol
(Smits and Riemann, 1988).
Results and Discussion
For this 10-day experiment in midsummer, the concentration of chlorophyll a was
about fivefold and phytoplankton primary production eightfold higher in -M than
in +M (Fig. 15.1). The abundance ofmter-feeding cladocerans and copepods was
about 50- and 4-fold higher in +M than in -M (Fig. 15.2). Mesozooplankton
clearance rate for phytoplankton was 3 LlL/day in +M and only about 0.03 LIL/
day in -M (Jeppesen et aI., unpublished data). As 90% and 60% of the chlorophyll
were in particles less than 20 Ilm in both enclosure types, most of the phytoplankton were presumably available for filter-feeding grazers. The difference in
chlorophyll a between +M and -M might be explained by zooplankton grazing,
although effects from macrophyte shading and high sedimentation rates cannot be
excluded.
The high mesozooplankton abundance in +M affected the abundance of the
protozooplankton. The densities of ciliates and flagellates in -M were about 200
and 2,100 cells/mI, respectively, which is 70- and 5-fold higher than in +M
(Fig. 15.2). Changes in zooplankton abundance and composition from July 25 to
August 1 were negligible, but a decrease in mesozooplankton abundance was
observed to take place after the introduction of planktivorous fish on August 1
(Jeppesen et al., unpublished data).
There was a three- to fourfold higher POC concentration in -M than in +M,
whereas no measurable difference in DOC concentrations was present at the start
of the experiment (Fig. 15.3). However, after about 6 days the concentration in +M
increased, whereas DOC decreased in -M after a peak value on July 29 (Fig. 15.3).
The DOC concentrations in +M and -M were significantly different except for the
two first samples (P <.001, paired t-test).
15. Bacterioplankton and Carbon Turnover
70
253
+. . . . . . . . . . -.. . . .
I
(/
60
50
E:'
0)
:::1.
......
co 40
~
..c:
Q.
...00
30
U
20
:E
10
-r pp
't'"............_.........1"". . . . . . . . . . . . .1'.........................,.....................,.....
Ol+------.--~~------r-----.------.-----+O
24-Jul
26-Jul
28-Jul
30-Jul
1994
01-Aug
03-Aug
05-Aug
Figure 15.1. Chlorophyll a (ChI) and primary production (PP) in enclosures with and
without submerged macrophytes. Means ± SD, n = 3 (three independent enclosures).
2140
4nn~------------------r----------
Macrophytes
o
m
----~
No macrophytes
Daphnia
_
Cladocerans
~ Copepods
Rotifers
_
CiliatesxE3
~
Flagellates x E3
Figure 15.2. Average meso· and protozooplankton abundance in enclosures with and
without submerged macrophytes. Averages for three enclosures and the 9·day experimental
period. Data on ciliates are provided by Klaus Jiirgens.
254
M. S0ndergaard et al.
10,--------------------------------------,
-o
8
EO
oa..
6
0:::::.
Cl
~
o
o
4
o
2
poe,
+M
o+------,,------,------,-------.------,------~
24-Jul
26-Jul
28-Jul
30-Jul
1994
01-Aug
03-Aug
05-Aug
Figure 15.3. Particulate (POC) and dissolved (DOC) organic carbon in enclosures with
and without submerged macrophytes. Means ± SD, n = 9 (triplicates in three enclosures).
The samples collected in a vertical profile at midday on July 31 and just before
sunrise on August 1 showed a tendency toward higher DOC concentrations just
above the sediment in both types of enclosures (Fig. 15.4). The profiles indicate a
DOC net efflux from the sediments. Furthermore, the DOC concentrations at night
were about 0.5-1 mg CIL higher than at midday.
The abundance of bacteria in -M was three- to fourfold higher than in +M
(Fig. 15.5A). The bacteria were initially small in all enclosures with an average
biovolume of 0.05-0.06 /lm3 • After July 31, the average cell volume in +M
increased to about 0.17 /lm3, whereas the cells in -M decreased to 0.039 /lm3 (but
see Jurgens and Jeppesen, this volume, Chapter 16). The mesozooplankton clearance rate for bacterioplankton in +M was 1.5 LIL/day (Jeppesen et aI., unpublished data). In -M, the mesozooplankton clearance rate for bacterioplankton
was 100-fold lower. As bacterial abundance during the experimental period did not
increase in +M and had a decreasing trend in -M, bacterial loss must at least have
equaled bacterial growth rates. The high mesozooplankton clearance rate of bacterioplankton in +M offers a plausible explanation for the loss in +M. The grazing
in -M was most probably dominated by flagellates, small ciliates, and rotifers.
Other factors such as virus and sedimentation of particle-associated bacteria also
contribute to the loss of bacterioplankton.
Despite the large difference in bacterial abundance between +M and -M, the
bacterial cell production did not differ markedly-although significantly--except
for July 27 (Fig. 15.5B). During the experimental period. the production decreased
255
15. Bacterioplankton and Carbon Turnover
70
E
-
~
c:
+ macrophytes
- macrophytes
60
~i9ht
50
i
,
(I)
E
I
40
'0
(I)
-J-
III
(I)
~ 30
.0
III
.c
a
\
20
(I)
0
•
'"
10
0
5
6
7
8
9
10
DOC (mg C/I)
Figure 15.4. Vertical profiles of DOC in enclosures with and without submerged macrophytes. Samples taken on July 31 (day) and August 1 before sunrise (night). Means ± SD, n
=3. Absent bars indicate SD within markers.
from about 8 to 2 x lOS cells/ml/h in both enclosure types (Fig. 15.5B). Initially,
the bacterial community in +M had a growth rate of 2A/day, decreasing to about
0.5/day. In -M the growth rate was about threefold lower and varied between 0.8
and 0.2/day. The reason(s) for the decrease in bacterial production during the
experiment is unknown, but it might be due to the exclusion of external substrate
sources. The generally higher specific growth rate in +M is in accordance with the
results presented in Jiirgens and Jeppesen (this volume, Chapter 16), where it is
shown that removal of zooplankton resulted in a substantial increase of bacteria.
The high grazing resulted in a higher specific growth rate, as can be predicted from
a logistic growth model, in which growth rates are high when the population
biomass is below the carrying capacity and will deminish as the carrying capacity
is approached (Wright, 1988). The cell production integrated for the entire experimental period was 91 ± 3 and 100 ± 14 x 106 cells/ml for -M and +M,
respectively (means ± SD, n = 3 enclosures of each type).
Bacterial Carbon Demand
The conversion of bacterial biovolume to carbon biomass is not trivial (see the
Materials and Methods section) and has major consequences for the values used to
assess the quantitative importance of bacteria. In -M, the bacteriop1ankton bio-
M. S¢ndergaard et al.
256
30~-------------------------------------
A
~
~
(lJ
()
CD
~
25
- macrophytes
20
(lJ
()
1ij
"0
15
c
:::J
.0
C\l
ro
10
·c
(lJ
t5
&l
5
o+-------------------------------------~
B
. + macrophytes
E 12
-....
.!!J.
Qj
()
L{)
~
9
c
o
t5:::J
"0
o
6
L-
a.
ro
·c
(lJ
t5
co
en
- macrophytes
3
O+------r-----,------,-----~-----,----~
24-Jul
26-Jul
28-Jul
30-Jul
1994
01-Aug
03-Aug
05-Aug
Figure 15.5. Bacterial abundance (A) and cell production (B) in enclosures with and
without submerged macrophytes. Means ± SD, n = 9 (triplicates in three enclosures).
Absent bars indicate SD within markers.
mass decreased from 160 to 80 )lg CIL and from 400 to 270 )lg CIL using 100 fg
(scenario I) and C = 120 X V0 72 (scenario 11), respectively. In +M, the
biomass increased from 40 to 120 )lg CIL in scenario I and from 140 to 210 )lg CIL
in scenario II. The use of a scaling factor with respect to cell size reduced the
relative variations over time and the difference between -M and +M.
Bacterial carbon demand (Bd was calculated according to
C/)lm3
257
15. Bacterioplankton and Carbon Turnover
Table 15.1. Carbon Flow Scenarios in Enclosures in Lake Stigsholma
Variables/substrate
sources
EOC
Sz
Be
Sb
DOCT
C s+p
EOC
Sz
Be
Sb
DOCT
Cs
Scenario I (100 fg C/~m3)
With macrophytes
9
40
160
-110
140
250
Without macrophytes
80
22 (82)
94
8 (68)
-120
0
Scenario II (C = 120 X YO.72)
9
85
430
-335
140
475
80
50 (110)
270
-140 (-80)
-120
20(0)
aAverage values of Ilg CIL/day calculated from values integrated over 9 days. EOC, extracellular
organic release by phytoplankton; Sz, bacterial substrate from mesozooplankton grazing and substrate
production by rnicro- and proto zooplankton grazing on phytoplankton in brackets; Be, bacterial
carbon demand; Sb, surplus or deficit of planktonic substrate production for bacteria; DOCT, changes
in the concentration of DOC; Cs+p, carbon from sediment (s) and plants (p) to balance the demand.
Assumptions used to calculate the two carbon flow scenarios:
I. Primary production (PP) in +M and -M: 100 and 900 Jlg CIL/day, respectively.
2. Algal respiration (R): 10% of PP.
3. EOC: 10% of (PP - R).
4. POC: 270 Jlg C/Uday and 0 in -M and +M, respectively.
5. DOC: 140 Jlg CIL/day in +M and-120 Jlg C/Uday in-M.
6. Zooplankton substrate recycling: 30% of ingestion (Hygum et aI., 1997). The total community
grazing was calculated from the production measurements compensated for changes in biomasses
(e.g., the grazing on bacteria in -M was calculated as bacterial production plus the decrease in biomass
and phytoplankton grazing in +M was subtracted the average increase in biomass). Phytoplankton
grazing in -M was attributed to zooplankton <50 Jlffi but allowing 50% for sedimentation. As grazing
by rnicrozooplankton was not measured, their potential carbon recycling is presented in brackets.
7. Bacterial biomass: 10 Jlg C/Uday in +M (both scenarios) and -10 or -18 Jlg C/Uday in -M, for
100 fg C/llm3 and C = 120 X YJ.72, respectively.
8. Bacterial growth yield: 50% from bacterial regrowth experiments (Sf/lndergaard and TheilNielsen, 1997).
Be = y-I . BB . ~
where Y = carbon growth yield, BB = carbon biomass, and 1..1. = growth rate. The
integrated production over the experimental period was 2.8-fold higher in scenario
II than in I for both +M and -M. In both scenarios, the bacterial production was
1.7-fold higher in +M than in -M (Table 15.1).
258
M. S!ilndergaard et al.
Planktonic Carbon Flow Scenarios
The measurements of carbon pool changes, phytoplankton production, clearance
rates, and planktonic bacterial production made it possible to construct a tentative
budget for planktonic bacterial carbon demand and substrate sources. The values
are calculated assuming carbon transfer from particles to bacterioplankton via
phytoplankton extracellular release (EOe) and zooplankton grazing on both
phyto- and bacterioplankton. Bacterial use of the ambient DOC pool is accounted
for as the changes in DOC enter the calculations. The scenarios in Table 15.1 are
based on the two "selected" calculations of bacterial biomass and a series of
assumptions concerning the planktonic flow of carbon to bacteria (see Table 15.1
footnote).
The consequences of the two scenarios are quantitative differences in bacterial
carbon demand, rates of grazer-recycled carbon, and EOe (Table 15.1). With
macrophytes present, both scenarios resulted in a substrate "deficit" for the bacterioplankton of either 110 or 335 J..lg elL/day (Table 15.1). To these deficits
should be added 140 J..lg elL/day caused by the increase in DOC. The sediment and
the macrophyte-periphyton complex are the only sources to cover the deficit as
other influxes can be considered negligible and the concentration of DOC increased.
Release of EOe and grazer-produced substrate balanced the bacterioplankton
carbon demand in -M and scenario 1. However, the decrease in DOC added 120
J..lg e/L/day, resulting in a substrate surplus (Table 15.1). In -M and scenario II, the
pelagic deficit was 140 J..lg elL/day, but with the decrease in DOC, only a minor
substrate deficit of 20 J..lg elL/day had to be accounted for by a net efflux from the
sediment. Including grazing of phytoplankton by the micrograzers enhanced zooplankton organic recycling to a quantity in which the bacterioplankton carbon
demand could be accounted for by the planktonic processes.
These theoretical scenarios are open to criticism. The conclusion that the
bacterioplankton in +M had a much larger apparent deficit for their carbon demand than in -M is very robust as long as identical assumptions concerning
carbon demand, EOe release, and organic recycling are used for both +M and -M.
We have chosen low phytoplankton respiration and rather high EOe and recycling
values (see Table 15.1 footnote), which all act to increase the planktonic carbon
flux to bacteria and thus diminish the suggested influence from macrophytes and
sediment.
Aside from bacterial biomass, the calculated bacterial carbon demand is influenced by the choice of the conversion factor for cell production per mole
thymidine incorporated and the growth yield. We have no reason to expect the
conversion factor to differ between -M and +M, and bacterial growth experiments
with water from both +M and -M showed no difference in growth yield (S!ilndergaard and Theil-Nielsen, 1997).
The measured cell sizes are critical for the calculation of bacterial biomass and
carbon production. If large cells have escaped our measurements in -M, the
calculated differences between +M and -M will diminish and eventually disap-
15. Bacterioplankton and Carbon Turnover
259
pear. Similar production values would be reached if the average cell volume for
growing cells in -M was about threefold larger than the measured average. In a
study on grazing-resistant bacteria in Lake Stigsholm, but using a different counting and sizing procedure than the present, Jiirgens and Jeppesen (this volume,
Chapter 16) reported lower abundances and generally larger cells than we have
measured. Although the use of their biovolumes resulted in a diminished difference in bacterioplankton carbon demand between -M and +M, an organic input
by the macrophyte-periphyton complex was still needed to establish a mass
balance, and our general conclusions were not invalidated.
Conclusions
It is demonstrated that the presence of submerged macrophytes had a profound
effect on the structure of the pelagic biota. The abundance of ciliates, flagellates,
and bacterioplankton in +M was low compared with -M, which had a fivefold
higher concentration of phytoplankton. The high mesozooplankton clearance rate
in +M was one reason for the difference, although other causes should not be
neglected. With respect to bacterioplankton production, a significantly (P <.00 I,
paired t-test) higher growth rate in +M in six of eight measurements compensated
for the lower cell abundance and resulted in similar cell production rates. Bacterioplankton production did not relate to phytoplankton biomass and production
in these shallow lake enclosures. It should be emphasized that a high bacterioplankton productivity was achieved in both enclosure types, which had very
different biological structures controlled by the absence or presence of a dense
population of submerged macrophytes.
A high bacterioplankton production was measured in +M despite a low phytoplankton biomass, and their carbon demand could only be balanced assuming the
macrophyte-periphyton complex as a substrate source. Total bacterial production
in the enclosures does not only include bacterioplankton but also the bacterial
production taking place on or in the sediment and in the biofilm on the macrophytes. The production in the sediments is unknown, but bacterial production on
Potamogeton pectinatus is high (Theil-Nielsen and S0ndergaard, unpublished data)
and adds to the total bacterial carbon demand in +M. This reinforces the position of
macrophytes as an active component in the DOC dynamics and metabolic activity of
shallow lakes and their impact on the pelagic microbial food web.
Acknowledgments. The comments and suggestions by Bob Wetzel and Klaus Jiirgens
and the technical skills of Gitte Jacobsen and Nils Willumsen are appreciated. This
study was supported by the Danish Environmental Research Programme.
References
Carlson, C.A.; Ducklow, H.W. Growth of bacterioplankton and consumption of dissolved
organic carbon in the Sargasso Sea. Aquat. Microb. Ecol. 10:69-85; 1996.
260
M. S¢ndergaard et al.
Carpenter, S.R.; Kitchell, J.F., eds. The trophic cascade in lakes. Cambridge: Cambridge
University Press; 1993.
Christoffersen, K.; Riemann, B.; Klysner, A.; S¢ndergaard, M. Potential role of zooplankton in structuring a plankton community in eutrophic lake water. Limnol. Oceanogr. 38:561-573; 1993.
Coveney, M.F.; Wetzel, R.G. Biomass, production, and specific growth rate of bacterioplankton and coupling to phytoplankton in an oligotrophic lake. Limnol. Oceanogr.
40:1187-1200; 1995.
del Giorgio, P.A.; Peters, R.H. Patterns in planktonic P:R ratios in lakes: influence of lake
trophy and dissolved organic carbon. Limnol. Oceanogr. 39:772-787; 1994.
Fagerbakke, K.M.; Heldal, M.; Norland, S. Content of carbon, nitrogen, oxygen, sulfur and
phosphorus in native and cultured bacteria. Aquat. Microb. Ecol. 10: 15-27; 1996.
Findlay, S.; Pace, M.L.; Lints, D.; Howe, K. Bacterial metabolism of organic carbon in the
tidal freshwater Hudson Estuary. Mar. Ecol. Prog. Ser. 89: 147-153; 1992.
Fuhrman, J.A.; Azam, F. Bacterioplankton secondary production estimates for coastal
waters of British Columbia, Antarctica, and California. Appl. Environ. Microbiol. 39:
1085-1095; 1980.
Hessen, D.O. Dissolved organic carbon in a humic lake: effects on bacterial production and
respiration. Hydrobiologia 229: 115-123; 1992.
Hygum, B.; Petersen, J.W.; S¢ndergaard, M. Dissolved organic carbon release by zooplankton grazing activity-a high quality substrate pool for bacteria. J. Plankton Res.
19:97-1l1; 1997.
Jeppesen, E.; Sortkj:er, 0.; S¢ndergaard, M.; Erlandsen, M. Impact of a trophic cascade on
heterotrophic bacterioplankton production in two shallow fish-manipulated lakes. Arch.
Hydrobiol. Beih. Ergebn. Limnol. 37:219-231; 1992.
Kairesalo, T.; Lehtovaara, A.; Saukkonen, P. Littoral-pelagial interchange and the decomposition of dissolved organic matter in a poly humic lake. Hydrobiologia 229: 199-224;
1992.
Kitchell, J.F.; Carpenter, S.R. Synthesis and new directions. In: Carpenter, S.R.; Kitchell,
J.F., eds. The trophic cascade in lakes. Cambridge: Cambridge University Press; 1993:
332-350.
Norland, S. The relationship between biomass and volume of bacteria. In: Kemp, P.F.;
Sherr, B.F.; Sherr, E.B.; Cole, J.J., eds. Handbook of methods in aquatic microbial
ecology. Boca Raton, FL: Lewis Publ.; 1993:303-307.
Pace, M.L. Heterotrophic microbial processes. In: Carpenter, S.R.; Kitchell, J.F., eds. The
trophic cascade in lakes. Cambridge: Cambridge University Press; 1993:252-277.
Porter, K.; Feig, Y.S. The use of DAPI for identifying and counting aquatic microflora.
Limnol. Oceanogr. 25:943-948; 1980.
Riemann, B.; Christoffersen, K. Microbial trophodynamics in temperate lakes. Mar. Microb.
Food Webs 7:69-100; 1993.
SchlUter, L.; Riemann, 8.; S¢ndergaard, M. Nutrient limitation in relation to phytoplankton
carotenoid/chlorophyll a ratios in freshwater mesocosms. J. Plankton Res. 19:891-906;
1997.
Simek, K.; Macek, M.; Pernthaler, 1.; Straskrabova, v.; Psenner, R. Can freshwater planktonic ciliates survive on a diet ofpicoplankton? 1. Plankton Res. 18:597-613; 1996.
Simon, M.; Azam, F. Protein content and protein synthesis rates of planktonic marine
bacteria. Mar. Ecol. Prog. Ser. 51 :201-213; 1989.
Smits, 1.; Riemann, B. Cell production derived from 3-H-thymidine incorporation using
freshwater bacteria. Appl. Environ. Microbiol. 54:2213-2219; 1988.
S¢ndergaard, M.; Middelboe, M. Measurements of particulate organic carbon: a note on the
use of glass fiber (GFIF) and Anodisc filters. Arch. Hydrobiol. 127:73-85; 1993.
S¢ndergaard, M.; Hansen, B.; Markager, S. Dynamics of dissolved organic carbon lability
during a clear water phase in a eutrophic lake. Limnol. Oceanogr. 40:46-54; 1995.
15. Bacterioplankton and Carbon Turnover
261
S!2!ndergaard, M.; Theil-Nielsen, J. Bacterial growth efficiency in lake water cultures.
Aquat. Microb. Ecol. 12:115-122; 1997.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Tranvik, L. Allochthonous dissolved organic matter as an energy source for pelagic bacteria
and the concept of the microbial loop. Hydrobiologia 229: 107-114; 1992.
Wetzel, R.G. Gradient-dominated ecosystems: sources and regulatory functions of dissolved organic matter in freshwater ecosystems. Hydrobiologia 229: 181-198; 1992.
Wright, R.T. Methods for evaluating the interaction of substrate and grazing as factors
controlling planktonic bacteria. Arch. Hydrobiol. Beih. Ergebn. Limnol. 31 :229-242;
1988.
16.
Cascading Effects on Microbial Food Web
Structure in a Dense Macrophyte Bed
Klaus JUrgens and Erik Jeppesen
Introduction
Heterotrophic microorganisms playa major role in the carbon and energy flow and
nutrient recycling of aquatic systems. Planktonic bacteria are regulated by the
supply of organic and inorganic nutrients and by predation of bacterivorous
organisms. Most studies on controlling mechanisms of bacterioplankton have
focused on assessing the direct effects of these factors. However, more recent
studies have revealed that microbial and classic food webs have an array of
interdependencies and are linked in many different ways (Turner and Roff, 1993).
Therefore, the structure of the whole planktonic community must be considered
for a better understanding of population dynamics at the microbial level (Pace et
al., 1990).
Cascading trophic interactions are known to play an important role in the
transfer of fish predation effects via zooplankton to the phytoplankton community
(Carpenter et aI., 1985). The trophic cascade concept has recently been examined
with respect to the effects of zooplankton on the microbial food web (Pace, 1993;
JUrgens et aI., 1994). The capability of certain zooplankton groups (e.g., daphnids)
to consume bacteria and those predation patterns on protozoans are key links
between the classic and microbial food webs. Different groups of metazooplankton such as daphnids, copepods, and rotifers have strong group-specific
impacts on planktonic protozoans (e.g., Arndt, 1993; Jiirgens, 1994; Sanders et al.,
1994). The protozoan community structure is to some extent a reflection of the
262
16. Cascading Effects on Microbial Food Web Structure
263
metazooplankton predation regime. If and how these effects on the protozoan
community are transferred to the bacterial level is not clear yet. Predatory cascades from zooplankton to bacteria could not be detected in enclosure experiments
and whole-lake manipulations in an oligotrophic lake (Pace, 1993). By contrast,
drastic cascading effects after zooplankton manipulation, which strongly altered
bacterial biomass and community structure, were found in an enclosure experiment in a mesotrophic lake (Jiirgens et aI., 1994).
Submerged macrophytes have a major impact on the biological structure and
water quality of shallow lakes (e.g., Scheffer et aI., 1993). Dense macrophyte beds
can act as a refuge for large zooplankton species, which in turn control the
phytoplankton (Timms and Moss, 1984; Schriver et aI., 1995; Lauridsen and
Buenk, 1996), and this may affect also the structure and function of the microbial
community. Microbial food web structure and dynamics were part of an integrated
study of the impact of macrophytes on the biological structure in a shallow
eutrophic lake (Jeppesen et aI., submitted). Bacterial production and coupling to
dissolved organic carbon (DOC) from the same experimental system are reported
by S~ndergaard et al. (this volume, Chapter 15). Here, we compare the composition and structure of protozoan and bacterial populations inside and outside submerged macrophyte beds. Results from fractionation experiments revealed the
importance of zooplankton predation as a determining factor for the microbial
food web structure.
Material and Methods
Lake Stigsholm is a shallow eutrophic lake in central Jutland, Denmark (mean
depth, 0.8 m; mean total Pconcentration, 0.15 mg/L). During the summer of 1994,
a large part of the littoral zone was covered by dense vegetation of submerged
macrophytes, mainly Potamogeton pectinatus. A detailed experimental design of
the enclosure system is given in Jeppesen et ai. (submitted). Briefly, six 5-m
diameter patches (three replicates each) were either kept free from plant vegetation (M-) or macrophytes were allowed to grow undisturbed (M+, plant infested
volume 40-55%). To assess the plankton community in these patches, without
interference from surrounding water, vertical polyethylene sheets were used to
enclose the patches for a 2-week period. All fish inside the enclosures were
removed by electrofishing and trapping. After 1 week, planktivorous perch were
again added to the enclosures. Sampling of organisms began 2 days after the
enclosement.
For a detailed examination of microbial food web structure and for size-fractionation experiments, we have used only two of the enclosures, in the following
referred to as M- (no vegetation) and M+ (with macrophytes). Twice during the
enclosement, we performed size-fractionation experiments; one (experiment 1)
under fish-free conditions 3 days after enclosement and the second (experiment 2)
2 days after restocking of fish. Water from the enclosures was either left unfiltered
(UF) or filtered through a 20-llm mesh net to remove both rneso- and most of the
264
K. Jiirgens and E. Jeppesen
microzooplankton. By comparing the population developments of the different
microbial components with and without zooplankton, this design should yield
information on the predation impact of the zooplankton communities in M+ and
M-.
After filling the water into transparent 4.8-L Nalgene polycarbonate bottles,
these were fixed to the enclosure frame, approximately 30 cm below the surface.
Samples for counting of organisms were taken at least once a day. Lugol-fixed
samples were used for counting of ciliates. For determination of the dominant
species, some selected samples were postfixed with Bouins fixative, and ciliate
species were determined after impregnation with Protargol according to Skibbe
(1994). Formalin-fixed samples were used for enumeration of bacteria, heterotrophic nanoflagellates (HNF), and autotrophic picoplankton (APP) after DAPI
(4'6-diarnidine-2-phenyl-indole; Sigma) staining (Porter and Feig, 1980). Volumes
of bacteria and APP were determined with an image analysis system (SIS GmbH,
Mlinster, Germany) connected to an epifluorescence microscope. We considered
the following bacterial morphologies as (protozan-) grazing-resistant bacteria
(GRB, see Jtirgens and Glide, 1994): long filamentous bacteria (cells or chains
approximately 7 /lm in length) and aggregated bacteria (particles with attached
bacteria or purely bacterial aggregates with a diameter of 10 /lm). GRB were
directly counted and measured with an ocular grid from the same DAPI preparations as total bacteria.
Results and Discussion
The enclosement of littoral patches with and without macrophytes enabled a
detailed analysis of the planktonic community. General composition and development of planktonic organisms proved to be very similar in the replicate enclosures
both with and without macrophytes (Jeppesen et al., submitted). The separation
from the surrounding water, especially from fish predation, might have led to some
changes in the regulating mechanisms of the enclosed organisms. However, the
relative stability of the plankton community structure during the 2-week study
period (data in Jeppesen et al., submitted) indicated that the enclosure did not
significantly alter the biological structure. Therefore, the experimental data characterize the plankton community that had previously developed within the macrophyte bed or in areas without submerged vegetation, respectively.
The decisive factor for the structure of the microbial food web was that the
zooplankton composition was different (Table 16.1): a high density of different
cladocerans and cyclopoid copepods was found in M+, whereas ciliates,
rotifers, and cyclopoid cope pods dominated in M-. The zooplankton communities are the well-known result of different degrees of fish predation:
large-bodied meso zooplankton were able to survive in the macrophytes probably
due to the refuge effect, and small-bodied zooplankton, mainly microzooplankton, developed without macrophytes, indicating an intense fish predation
pressure.
16. Cascading Effects on Microbial Food Web Structure
265
Table 16.1. Density ofImportant Planktonic Organisms in Enclosures with (M+) and without
(M-) Macrophytes at the Beginning of the Size-Fractionation Experimentsa
Experiment 1
M+
Pico-/nanoplankton
2.8 ±0.5
Total bacteria (10 6/ml)
Filamentsb (103/ml)
13.8 ± 3.9
AggregatesC (103/ml)
3.7 ±0.5
APP (l05Imi)
0.2±0.1
HNF (103/ml)
0.8 ±0.6
Micro-Imacrozooplankton
Ciliates (/mL)
1.5±0.4
Rotifers (IL)
436±26
Daphnids (IL)
148 ± 13
Other cladocerans (IL)
213 ± 104
Cyclopoid copepods (IL) 1,184 ± 331
Phytoplankton
Chlorophyll a (llglL)
12.0± 5.3
Experiment 2
M-
M+
M-
14.1 ±0.5
31.8 ± 2.8
4.2 ±2.3
5.8 ± 1.6
5.5 ± 1.4
4.0 ±O.4
8.5 ± 1.5
2.1 ±O.7
0.7 ±0.2
0.9 ± 0.4
8.9 ± 0.1
133.4± 13.8
3.9 ±0.9
25.5 ±0.7
6.5 ±0.1
245 ± 52
15,140 ± 4,845
1.5±0.4
1.5 ± 1.0
301 ± 3
11.7 ±9.0
199 ± 94
61 ±30
247 ± 103
635 ±298
216 ± 105
20,095 ± 5,024
0
5 ± 1.7
418 ± 176
60.3 ± 6.1
15.0±6.1
67.3 ± 19.6
QMeans ± SD of three replicate enclosures (zooplankton) or means ± SD of three replicate treatments from
one enclosure (pico-/nanoplankton).
bpilamentous bacteria >6 ~m.
C Aggregates> I 0 /.lm with attached bacteria.
Tremendous differences between M- and M+ were also visible in the structure
of the microbial food web. In M+, the abundance of phytoplankton, protozoans,
and bacteria was very low, whereas in M- an abundant and diverse assemblage of
auto- and heterotrophic pico-, nano-, and microplankton coexisted with the metazooplankton community (Tables 16.1 and 16.2). Ciliates represented the most
drastic difference in the microbial food web (Table 16.2). Ciliate density was
approximately two orders of magnitude higher in M- compared with M+, and
species composition differed. Some of the ciliates found in M+, such as the
hypotrichous species, are benthic, probably macrophyte-associated species. Truly
planktonic forms clearly dominated in M-. Different trophic levels and feeding
modes (bacterivorous, algivorous, and carnivorous species) were present within
the ciliate community. The most abundant groups were scuticociliates (mainly
bacterial feeders) and small oligotrichous and prostomatid species, which feed
mainly on nanophytoplankton (Miiller et aI., 1991). But larger predatory ciliates
such as Urotricha pelagica and species from the Haptorida order were also present
in higher numbers (up to 25 individuals/rnl) in M-. From the different cell sizes,
we assume that different functional groups were probably also present among the
heterotrophic flagellates in M-: small bacterivorous species (3-5 Ilm) and larger
mainly algivorous species (10-15 Ilm). The diverse protozoan species composition in M- must be exerting a significant grazing pressure on the bacterial as well
266
K. Jurgens and E. Jeppesen
Table 16.2. Ciliate Species Composition in Enclosures with (M+) and without (M-)
Macrophytes
Ciliate taxa
Total abundance" (per m!)
Prostomatida
Urotriclw cf. furcatalfarcta
Urotriclw cf. pelagica
Coleps sp.
Oligotrichida
Halteria grandin ella
Strobilidium lacustris
Strobilidium sp. «25 I!m)
Haptorida
Askenasia sp.
Monodinium sp.
Lagynophria sp.
Enchylis sp.
Actinobolina sp.
Scuticociliatida
Cyclidium sp.
Cinetochilum margaritaceum
Peritrichida
Vorticella spp.
Vorticella mayeri
Colpodea
Cyrtolophosis mucicola
Hypotrichia
Aspidisca sp.
Oxytricha sp.
M+
M-
3.8 ± 3.2
282 ±79
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
aMean ± SD from 2 weeks.
as on the algal community. Further, many predator-prey interactions probably
occur between different protozoan groups.
Differences between M+ and M- occurred at the bacterial level as well:
bacterial concentrations were four- to fivefold higher in M- when compared with
M+ and exhibited a more pronounced morphological diversity with a high proportion of filamentous forms and aggregated bacteria (Fig. 16.1). These complex cell
morphologies are resistant to protozoan grazing and generally appear when predation pressure by protozoans is high, especially during HNF population peaks
(Jurgens and Gude, 1994). APP, here exclusively chroococcid cyanobacteria,
constituted another important part of the picoplankton in M- but were at a low
level in M+. In M-, their contribution to total picoplankton biomass even exceeded heterotrophic bacteria in the second experiment (Table 16.3). Also,
16. Cascading Effects on Microbial Food Web Structure
267
Figure 16.1. Epifluorescence microphotographs of DAPI-stained preparations (1 ml of
sample on 0.2-I.lm Nucleporefilter). (A) M+, start of experiment 2; (B) M-, start of
experiment 2; (C) M+, 20 Ilm filtered, after 3 days. Scale bar = 10 Ilm.
268
K. Jiirgens and E. Jeppesen
Table 16.3. Estimated Biovolumes (105 J..Lm3/ml) of Picoplankton in M+ and M- at the
Beginning of Experiments 1 and 2a
M-
Bacteria
Filaments
APP
M+
Experiment 1
Experiment 2
Experiment 1
Experiment 2
8.34±0.28
0.39 ±0.06
6.22 ± 1.25
5.34±0.08
2.64 ±0.27
28.05 ±0.78
2.52± 0.31
0.13 ± 0.03
0.22 ± 0.06
3.61 ±0.85
0.12 ±0.02
0.80 ±0.37
aMeans ± so of three replicate treatments. Autotrophic picoplankton (APP; mean cell volume,
l.lO !lm\ bacteria (freely suspended rods and cocci, mean cell volume 0.06 !lm3 and 0.09 !lm3 for
M- and M+, respectively), and filaments (filamentous bacteria, mean cell volume 1.0-2.0 !lm\
filamentous bacteria contributed substantially to the bacterial biomass in M(Table 16.3).
The size-fractionation experiments revealed that the impacts of zooplankton
greater than 20 Ilm on the microbial food web differed between M+ and M-. We
considered top-down effects of zooplankton to be effective when the population
levels of the various microbial components differed significantly between UF and
less than 20 Ilm filtered water after 24 hours. The results of the first experiment are
shown in Figure 16.2; the statistical results of both experiments, which showed
virtually similar trends, are summarized in Table 16.4. The removal of zooplankton in M+ resulted in an immediate increase in bacteria and protozoans
compared with the UF controls. Most pronounced was the response of heterotrophic bacteria and HNF; within 24 hours, their abundance increased by factors
of 3 and 9, respectively. Chlorophyll a increased by a factor of 4 within 4 days.
These findings are in accordance with the general notion that Daphnia, which
occurred in high densities in M+, can control bacteria, algae, and a wide range of
protozoans (Christoffersen et aI., 1993; Jurgens, 1994).
Removal of zooplankton greater than 20 Ilm in M- had little or no effect on
pico- and nanoplankton. Population levels remained constant or increased, which
was similar to the results for the UF controls (Fig. 16.1), with the only significant
increase being that of filamentous bacteria (Table 16.4). We assume that complex
predatory interactions within the microbial food web took place in M-, which did
not become visible in our experimental design because they prevailed in the less
than 20-llm fraction. It has previously been shown by size-fractionations with
smaller filter pore sizes that several trophic links can exist even among organisms
less than 10 Ilm (Wikner and Hagstrom, 1988).
A pronounced microbial succession became visible in M+ after removal of
zooplankton and incubation of the bottles for 5 days (Fig. 16.3). The immediate
peak in bacteria (within 12 hours) was followed by an increase in HNF, which was
probably responsible for the following decline in bacterial concentrations. APP
increased continuously during the experiment but at a much lower rate than
heterotrophic bacteria. Also ciliates showed an exponential increase during the
269
16. Cascading Effects on Microbial Food Web Structure
M-
M+
15
15
10
10
5
5
0
0
E 0.8
8
6
4
E
'"0
:!:.
.~
Q;
ti
til
OJ
"b
.....
a.. 0.4
a..
«
E
0
'".....
u..
z
r
I
0.0
12
2
0
12
9
9
6
6
3
3
0
40
0
30
300
200
~ 10
100
.2l
C:3
0
120
E
'"0
..... 90
0
120
VI
60
60
E
30
30
u:::
0
0
80
80
60
60
C
Q)
~
-Ol
~
til
:cu
I
400
20
VI
II
I to
~tl
90
40
40
20
20
0
0
UF
<20
UF
<20
Figure 16.2. Size-fractionation (experiment 1). Development of bacteria, APP, HNF, ciliates, filamentous bacteria, and chlorophyll a in unfiltered (UF) and 20 Ilm filtered water
from enclosure M+ and M-. t] is 24 hours except for chlorophyll a (chi a), where t] is
4 days. Note difference in scales for APP and ciliates.
270
K. Jiirgens and E. Jeppesen
Table 16.4. Statistical Comparison of the Means (after 24 hours) of Unfiltered (UF) and
20 ).lm Filtered Water for Enclosures M+ and M- and for Experiments 1 and 2a
Experiment 1
Organisms
Bacteria
APP
Filaments
HNF
Ciliates
Experiment 2
M+
M-
M+
M-
<0.0001
<0.05
<0.001
<0.0025
<0.05
ns
ns
<0.05
ns
ns
<0.005
<0.05
<0.01
<0.05
<0.05
ns
ns
<0.05
ns
ns
aProbability (P) levels are from unpaired two-sample I-tests. ns, difference of means not significant
at the 0.05 level.
incubation, with Halteria grandinella as the dominant species (80% of total
abundance). The decline in HNF might be a result of both ciliate predation and
depletion of food resources. Grazing-resistant bacteria (filamentous forms and
aggregates) increased as well during the incubation period. This succession sequence is very similar to the one reported by Jiirgens et aI. (1994) from an
enclosure study in which removal of Daphnia resulted in an increase in phagotrophic protozoans, which shifted the bacterial assemblage toward protozoaninedible forms.
Other aspects of microbial interactions can be deduced from this succession in
fractionated water from M+. After removal of zooplankton, a huge bacterial
production, which was previously controlled by mesozooplankton, became visible. Later in the experiment, it was controlled by HNF. The increase in bacteria
within the first 12 hours after zooplankton removal corresponds to a bacterial
doubling time of 6.6 hours. This is in agreement with direct measurements of
bacterial production in the enclosures (S~ndergaard et aI., this volume, Chapter 15), which showed that bacteria in M+ were about as productive as in M-,
despite low concentrations of phytoplankton and a substantially lower bacterial
abundance.
APP seemed to have a slightly different role than heterotrophic bacteria in the
microbial food web of Lake Stigsholm. Their exponential increase after zooplankton removal in M+ (but at a lower growth rate than bacteria) revealed the
efficient top-down control by mesozooplankton. In contrast to bacteria, APP were
not reduced after HNF and ciliates achieved higher numbers in the incubations
(Fig. 16.3). Together with the fact that APP were present in high densities in the
M- enclosures, this might indicate that they were to a lesser extent subject to
protozoan predation than were heterotrophic bacteria. This supports the view that
cyanobacteria have lower food quality and are probably negatively selected by
protozoans (e.g., Caron et al., 1991).
The size-fractionation experiments demonstrated that zooplankton predation in
M+ was the decisive factor for the observed microbial food web structure. Al-
1
271
16. Cascading Effects on Microbial Food Web Structure
15
E
co
0,...
Plcoplankton
12
I.
9
as
~
CD
6
I
I
lAPP
z
J:
CD
E
..!!l
u:
150
}J
9000
/
6000
- _--if'
./
/
GRB
40000
FIL
30000
Y"
--- -y---"
20000
~'L
100
§.
U)
CD
!
./
50000
c:
0
,...
~
Protozoans
0
Jg
4!
0
3000
-'E
5
3
12000
u.
6
3 n.
n.
2 «
0
'E
7
/1
--"
Jg
50
G
0
8000
6000 E
U)
CD
4000
iii
Cl
~
Cl
Cl
AGGR-
2000
10000
«
0
0
0
20
40
60
80
100
Time[h]
Figure 16.3. Development of picoplankton (bacteria, APP), heterotrophic protozoans (HNF,
ciliates), and grazing-resistant bacteria (ORB: aggregates, filaments) in 20 J.lIll filtered water of
enclosure M+ (experiment 1).
though the microbial community in the less than 20-f.,lm fraction is probably undergoing a transient state, it is interesting to note that after removal of zooplankton, bacteria
and protozoans reached similar levels as those in M-. Also, the general morphological
composition of bacteria became more similar (Fig. 16.1C).
The littoral patches, M+ and M-, were ideal for studying trophic interactions at
different food web constellations as the enclosures were situated in close proximity in
272
K. Jurgens and E. Jeppesen
the littoral zone of the same lake, and nutrient levels and water chemistry were
almost identical (Jeppesen et al., submitted). Although the macrophyte-periphyton
complex can release bacterial substrates and thus compensate for the lack of
phytoplankton in M+, this did not seem to be the crucial factor for the different
microbial food web structures. Instead, different zooplankton communities and
predatory cascades were of overwhelming importance.
The meso zooplankton community in M+ (dominated by cladocerans) exerted a
much stronger top-down control on pico- and nano-sized organisms than the
microzooplankton (and the copepods) in M-. The results are in general accordance
with the view that microzooplankton assemblages are much less able to control
nanophytoplankton than is a zooplankton community consisting of large filterfeeding species (e.g., Mazumder et aI., 1990). This is because the latter community
generally has a higher total biomass and exerts lower feeding selectivity (Lampert,
1988; Mazumder et a!., 1990). Picoplankton (bacteria, APP), exceeding a certain
cell size, probably find it harder to escape consumption by filter feeders such as
Daphnia but may develop some kind of resistance against predation by protozoans
(Jurgens and Gude, 1994). This is the reason why despite lower overall grazing
pressure on picoplankton in M- (and higher abundance), a larger proportion
consisted of protozoan-inedible or less digestible forms (filaments, aggregates,
cyanobacteria).
The different zooplankton communities of M- and M+ are a result of the previous
differing fish predation pressure, which was estimated to be very strong in M- but, due
to the refuge effect of the macrophytes, relatively low in M+ (Jeppesen et al.,
submitted). Consequently, predation effects can cascade from fish to the bacterial level
in these systems. It is probably a general feature of more eutrophic systems that
top-down regulation is a stronger shaping factor for the plankton community than food
resources (bottom-up regulation) (Jeppesen et al., 1997).
Our study and the comparable one by Jurgens et al. (1994) showed that this also
holds true for planktonic bacteria, which responded immediately to an alteration in
the predation regime. This is in contrast to Pace's model (1993), which assumed
that the main effects of the trophic cascade on bacteria are indirect, mediated through
alterations in phytoplankton biomass and productivity. We assume that this might be
the case in oligotrophic systems in which bacteria are more likely to be nutrient
limited, but direct predatory interactions that affect the structure and function of
microbial food webs are more important in systems of higher productivity.
Acknowledgments. We are grateful to Morten S!iSndergaard and Kirsten Christoffersen for their critical review of the manuscript and to Nancy Zehrbach, who
improved our wording. Special thanks go to Oliver Skibbe who introduced us to
Protargol staining and helped with ciliate species determination.
References
Arndt, H. Rotifers as predators on components of the microbial web (bacteria, heterotrophic
flagellates, ciliates)-a review. Hydrobiologia 255/256:231-246; 1993.
16. Cascading Effects on Microbial Food Web Structure
273
Caron, D.A.; Lim, E.L.; Miceli, G.; Waterbury, J.B.; Valois, EW. Grazing and utilization of
chroococcoid cyanobacteria and heterotrophic bacteria by protozoa in laboratory cultures and a coastal plankton community. Mar. Eco!. Prog. Ser. 76:205-217; 1991.
Carpenter, S.R.; Kitchell, J.E; Hodgson, J.R. Cascading trophic interactions and lake
productivity. BioScience 35:635--639; 1985.
Christoffersen, K; Riemann, B.; Klysner, A.; Spndergaard, M. Potential role offish predation and natural populations of zooplankton in structuring a plankton community in
eutrophic lake water. Limno!. Oceanogr. 38:561-573; 1993.
Jeppesen, E.; Jensen, J.P.; S¢ndergaard, M.; Lauridsen, T.; Pedersen, L.1.; Jensen, L.
Top-down control in freshwater lakes: the role of nutrient state, submerged macrophytes
and water depth. Hydrobiologia 342/343: 151-164; 1997.
Jeppesen, E.; S¢ndergaard, M.; S¢ndergaard, M.; Christoffersen, K; Jiirgens, K; TheilNielsen, J.; SchlUter, L. Cascading trophic interactions in the littoral zone of a shallow
lake (submitted).
Jiirgens, K. The impact of Daphnia on microbial food webs-a review. Mar. Microb. Food
Webs 8:295-324; 1994.
JUrgens, K; GUde, H. The potential importance of grazing-resistant bacteria in planktonic
systems. Mar. Eco!. Prog. Ser. 112: 169-188; 1994.
Jiirgens, K; Arndt, H.; Rothhaupt, KO. Zooplankton-mediated changes of bacterial community structure. Microb. Eco!. 27:27-42; 1994.
Lampert, W. The relationship between zooplankton biomass and grazing: a review. Limnologica 19: 11-20; 1988.
Lauridsen, T.; Buenk, I. Diel changes in the horizontal distribution of zooplankton in the
littoral zone of two shallow eutrophic lakes. Arch. Hydrobiol. 137: 161-176; 1996.
Mazumder, A.; McQueen, D.J.; Taylor, W.o.; Lean, D.R.S.; Dickman, M.D. Micro- and
mesozooplankton grazing on natural pico- and nanoplankton in contrasting plankton
communities produced by planktivore manipulation and fertilization. Arch. Hydrobio!.
118:257-282; 1990.
MUller, H.; Schone, A.; Pinto-Coelho, R.M.; Schweizer, A.; Weisse, T. Seasonal succession
of ciliates in Lake Constance. Microb. Eco!. 21:119-138; 1991.
Pace, ML Heterotrophic microbial processes. In: Carpenter, S.R.; Kitchell, J.P., eds. Cascading
trophic interactions. Cambridge: Cambridge University Press; 1993:252-277.
Pace, M.L.; McManus, G.B.; Findlay, S.E.G. Planktonic community structure determines the
fate of bacterial production in a temperate lake. Limno!. Oceanogr. 35:795-808; 1990.
Porter, KG.; Feig, Y.S. The use of DAPI for identifying and counting aquatic microflora.
Limnol. Oceanogr. 25:943-947; 1980.
Sanders, R.W.; Leeper, D.A.; King, e.H.; Porter, KG. Grazing by rotifers and crustacean
zooplankton on nanoplanktonic protists. Hydrobiologia 288: 167-181; 1994.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Eco!. Evo!. 8:275-279; 1993.
Schriver, P.; B¢gestrand, J.; Jeppesen, E.; S¢ndergaard, M. Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large-scale enclosure experiments in a shallow eutrophic lake. Freshwat. Bio!. 33:255-270; 1995.
Skibbe, O. An improved quantitative protargol stain for ciliates and other planktonic
protists. Arch. Hydrobiol. 130:339-347; 1994.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing in the presence of zooplanktivorous fish in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Turner, J.T.; Roff, J.e. Trophic levels and trophospecies in marine plankton: lessons from
the microbial food web. Mar. Microb. Food Webs 7:225-248; 1993.
Wikner, J.; Hagstrom, A. Evidence for a tightly coupled nanoplanktonic predator-prey link
regulating the bacterivores in the marine environment. Mar. Eco!. Prog. Ser. 50:137145; 1988.
17. Abundance, Size, and Growth of Heterotrophic
Nanoflagellates in Eutrophic Lakes with
Contrasting Daphnia and Macrophyte Densities
Kirsten Christoffersen
Introduction
Natural populations of heterotrophic nanoflagellates (HNF) can constitute an
important component of the nanoplankton community because they have high
growth rates and high predation rates on picoplankton (Riemann and Christoffersen, 1993). Seasonal variability in the HNF population (e.g., Carrick et aI., 1992)
has often been attributed to the abundance of picoplankton (especially bacteria).
Berninger et aI. (1991) established a predator-prey correlation based on data from
numerous lakes. Several other cross-system analyses have concluded that the
relationship is not strong and decreases in strength from oligotrophic to eutrophic
systems (Sanders et aI., 1992; Gasol and Vaque, 1993; Gasol, 1994).
The HNF biomass in eutrophic lakes is often small, relative to the biomass of
other microbes, during most of the year (Riemann and Christoffersen, 1993).
Numerous studies have shown that predation by ciliates, rotifers, and metazoans
can regulate HNF abundance and biomass in these ecosystems (e.g., Glide, 1988;
Sanders and Porter, 1990; Weisse, 1991; Arndt, 1993, 1994; Christoffersen et aI.,
1993; Jiirgens and Stolpe, 1995). The influence of cladocerans, in particular
Daphnia, on HNF has previously been described (Riemann, 1985; Vaque and
Pace, 1992). The food spectrum of Daphnia is large, and their ability to consume
nano-sized protozoans is widely recognized (Stoeckner and Capuzzo, 1990;
Arndt, 1993; Jiirgens, 1994). This implies that mechanisms operating on Daphnia
populations are also indirectly acting on the protozoan community.
274
17. Abundance, Size, and Growth of Nanoflagellates
275
In a recent study, Christoffersen et al. (1993) showed that the presence of
planktivorous fish changed the biomass and composition of zooplankton in a
eutrophic lake, and this, in tum, affected the microbial communities. When cladocerans dominated (Bosmina sp. and D. cucullata), they controlled the biomass of
phytoplankton, HNF, rotifers, and bacteria. However, when fish reduced the
cladoceran biomass, a microbial community developed despite the presence of
other metazoans (mainly copepods). An increasing number of studies have recognized the key position of Daphnia in freshwater systems, supporting the thesis that
a strong predator control is the structuring element in lakes (Jeppesen et al., 1992;
Riemann and Christoffersen, 1993; Jiirgens, 1994).
This study evaluates the specific interactions between populations of Daphnia
and HNF in eutrophic lakes on a cross-system basis and includes the potential
impact of macrophytes on these interactions. Because macrophytes act as refuges
for zooplankton under high fish predation (Schriver et al., 1995), it can be hypothesized that accumulation of zooplankton in macrophyte stands may lead to
intensive grazing activities influencing the microbial community. The data set
allowed also a test of the correlation between abundance of bacteria and HNF by
using the framework proposed by Gasol (1994) as well as a test of the effects of
temperature and bacterial abundance on HNF growth rates.
Material and Methods
The study was carried out in four shallow and eutrophic (total phosphorus, >0.1 mgIL
during summer) lakes: Frederiksborg Slots sill, Stigsholm sill, Ring sill, and Ramten sill (Table 17.1). The lakes had either naturally occurring submerged macrophytes or planted ones, except for Frederiksborg SlotsslII. Differences in
macrophyte density (i.e., plant-filled volume) within the specific lakes were obtained by manually adjusting the macrophytes either directly in the lake (Ring Sill)
or inside large enclosures (19.6 m2). The enclosures were constructed by rein-
Table 17.1. Some Characteristics of the Lakes Included in This Study
Trophic status
Frederiksborg
Slottss\'!
Stigsholm s\,!
Highly
eutrophic
Eutrophic
Ring s\,!
Highly
eutrophic
Eutrophic
Ramten s\,!
aGiven as relative plant-filled volume.
Mean
depth
Macrophytes
(densityt
Dominant
Daphnia species
3.1 m
None
D. cucullata
0.8m
Potanwgeton
(0-50%)
Potamogeton
(50-100%)
Ceratophyllum
(0-100%)
D. galeata,
D. cucullata
D. magna
2.1 m
l.Om
Daphnia sp.
276
K. Christoffersen
forced plastic and stainless steel or wooden poles (further details in Jeppesen et at,
submitted). Each enclosure had a diameter of 5 m and a water depth of 0.6-1.2 m.
All experiments were carried out between March-September during 1992-1995.
Water samples were collected daily (noon) or twice daily (noon and midnight)
during the experimental periods, and subsamples for identification and enumeration of HNF, bacteria, chlorophyll, and zooplankton were taken. The sampling
strategy is described in Sjljndergaard et al. (this volume, Chapter 15). HNF were
fixed from whole-water samples with glutaraldehyde (1.5% final concentration),
stained with DAPI within 24 hours and filtered onto 0.8-/..Im black Nuclepore or
Poretics filters. The filters were kept frozen (_20DC), and the number of flagellates
was later counted by using an epifluorescence microscope fitted with an ultraviolet filter set. Only cells of 2-10 j.1m in greatest dimension were recorded, and at
least 50 cells were counted from each filter. The HNF biovolume was determined
by measuring linear dimensions of 10-20 individuals per filter. Processing of
samples for abundance of bacteria and zooplankton is described in Sjljndergaard
et al. (this volume, Chapter 15).
Potential growth rates of HNF were measured under predator-free conditions
by incubation of prescreened water. Subsamples were produced from whole-water
samples by screening the water with gentle inverse filtration using a 25 L fractionator equipped with a nitex screen with a mesh size of 20 j.1m. The screened
water was added to l-L or 0.25-L acid-cleaned polycarbonate bottles and incubated in situ at 0.5-m depth for 24 hours. The bottles were mounted on a floating
rig and were thus exposed to natural water movements. An air bubble inside each
bottle ensured that the water was mixable. Calculations of growth rates were based
on total changes in abundance, assuming exponential growth.
The data set, including data on the abundance of bacteria, zooplankton, and
density of macrophytes, was analyzed by linear regression analysis (SAS JUMP)
of log-transformed (basee) data. In cases in which no Daphnia were found, each
value was increased by 1 (n+l).
A laboratory experiment was conducted to investigate the predation pressure of
Daphnia magna on natural HNF. Water from three stations (± macrophytes and
midlake) in Ring sjlj was prescreened through 20-j.1m nets and divided into 18
subsamples of 250 ml each. Fifteen D. magna (1.5 mm) were then added to each
of half of the samples, and all samples were incubated in the laboratory at in situ
temperature (25 DC) for 24 hours in dim light. The abundance of HNF was measured at the start and end of the incubation as described above. The daphnids were
collected from the lake and kept in prescreened (20 j.1m) lake water for 2 hours
before the experiment.
Results and Discussion
The abundance of HNF in the investigated lakes (Fig. 17.1) ranged from 0.1 x 102
cells/ml to 1.3 x 104/ml with marked seasonal variations. The abundance of
Daphnia covered a range from 0 to more than 1,OOOIL (Fig. 17.1). Different
277
17. Abundance, Size, and Growth of N anoflagellates
Figure 17.1. Relationship between the abundance of Daphnia sp. and heterotrophic nanoflagellates in four eutrophic and shallow
lakes. Measurements in replicated enclosures
or in the lakes are shown individually. Circles
denote systems with macrophytes, and triangles denote systems without macrophytes.
The ranges of macrophyte-filled volume in
the respective lakes are given in parentheses.
Five measurements at very high Daphnia
densities (512-1,342 DaphniaIL; 1-2 x 1<r
HNF/ml) following the general trend are
omitted for clarity.
15] Ring s. and Ramten I. (0-10011 m•• rophyte.'
L
E
3~.
000
..
00
0
~ 0.5
II)
{
0.1
1:] ~o0
g 0.5
.!!
Ql
.to.: . . t'" #"~.6.,t..
.6.
....
.....
...
..
.... " ..
..,
C
Sti~lhDlm s. {O-SOl m.cr.,~ytel'
.ott.
......
..
•
alto
.6.0
o~
ca
...
0..
10
0"
•
•
~ 15]: Frederiksborg Slotss. (no macrophytes)
e
'0
3
I
li; 0.5
;
:I:
0
0
0
0
OQ:
0
0
0.1
o
100
200
300
Daphnia abundance (1-1)
species of mainly medium-sized daphnids were present (Table 17.1), except in one
case in which the population was totally dominated by large D. magna (Ring s0).
The Daphnia populations were regulated by temperature and by fish predation
(data not presented here). Thus, situations with low densities of daphnids as, for
instance, the early spring or during intensive fish predation (typically during early
summer), were associated with higher abundances of HNF than when daphnids
were numerous (e.g., at high macrophyte density and/or low fish predation).
The macrophytes had no impact per se on the abundance of HNF (Fig. 17.1),
but when the macrophytes functioned as a refuge for Daphnia, they had a negative
effect on the popUlation size of HNF (compare Fig. 17.1 middle and lower panel).
This trend was not, however, found in Ring and Ramten S0 (Fig. 17.1 upper panel),
where the HNF abundance reached higher levels (1-3 x 103/ml) in the presence of
Daphnia than in the other two lakes. The fact that no enclosures were established
in this experiment implies that a continuous invasion ofHNF from the surrounding
macrophyte areas may have occurred and that this equaled the loss of HNF by
Daphnia due to grazing.
A correlation between Daphnia and HNF abundance showed a significant
negative relationship (Table 17.2). A general log-log model was used for this
relationship. Other types of data transformation (i.e., 1IHNF ;and log 11HNF) did
not improve the relationship. The biovolume of individual HNF cells was not
correlated with the abundance of Daphnia in the same manner (Table 17.2), albeit
a higher frequency of small HNF was found at high Daphnia abundance than at
low abundance. The explanation of these weak relationships is probably that a
variety of other organisms feeding on nano-sized plankton can develop when
Daphnia populations are low (Pace and Funke, 1991; Christoffersen et ai., 1993;
Jurgens, 1994). Consequently, a new correlation that only used Daphnia abundances higher than 0 individualsIL and the total HNF biovolume (i.e., abundance x
K. Christoffersen
278
Table 17.2. Summary of the Linear Regression Analysis on Log Tmnsfonned Data
(base,,)a
Log HNF (per ml) vs. log
Daphnia + 1 (per L)
Log HNF volume (/lm3/cell)
vs. log Daphnia + 1 (per L)
Log HNF total biovolume
(/lm3/ml) vs. log Daphnia
>0 (perL)
Log bacteria (per ml) vs.log
Daphnia + 1 (per L)
Log bacteria (per ml) vs. log
HNF(perml)
Log bacteria (per ml~ vs. log
HNF volume (/lm /eell)
Slope
(±SE)
Intercept
(±SE)
p
p
N
,2
251
0.172
-0.21 (0.03)
***
0.56 (0.09
***
251
0.100
-0.06 (0.012)
ns
3.08 (0.04)
***
187
0.210
-0.27 (0.03)
**
3.64 (0.10)
***
212
0.160
-0.11 (0.02)
***
2.52 (0.05)
***
253
0.002
-0.022 (0.03)
ns
2.21 (0.04)
***
253
0.000
0.0011 (0.03)
ns
2.20 (0.10)
***
aN, number of observations; ,2, correlation coefficient; ***p < .001; **p < .01; ns, not significant.
individual biovolume) was established (Fig. 17.2). This provided a slightly
stronger correlation (Table 17.2) and allowed the calculation that 0.33 DaphnialL
was needed to reduce the standing stock of HNF (in terms of total biovolume) by
50%. A similar low threshold of Daphnia abundance was found by J iirgens (1994)
when plotting the biomass of Daphnia versus the abundance of HNF. Daphnia
abundances are often above 10 individualslL in temperate eutrophic lakes during
summer (Riemann and Christoffersen, 1993). This implies that the growth rate of
1000
"I
0
'"E
8
E
00
::J.. 100
0
o
(I)
'"0
o
~
Q)
o
2
0
>
0
15
8'
E
10
0
0
0
0
Jl C@
0 0 0 0
0
0
0
'b 000
0
0
0
CXbc9 0
0Ol§>
o
(I)
00
00 0
0
0~c9
0
0
0
co (I) "Be
0000 0
~d'o
I
io,f
°d'o
0
o
02
0
00
00
000
0
0
z
0
00
00 0 0
0 0
00 0
0
00 0
0
Jl
cil
0
LL
0
o
OOoEP
2
20
Daphnia abundance
200
>
2000
0 (1-1)
Figure 17.2. Relationship between the abundance of Daphnia and the total HNF biovolume. Note that both axes are on a log scale.
279
17. Abundance, Size, and Growth of N anoflagellates
Figure 17.3. Laboratory experiment
showing the effect of 0 and 15 D. magna
on the growth rates of natural populations of heterotrophic nanoflagellates
(from three locations in Ring s!2S) after
a 24-hour incubation. Each column
represents the average of triplicates,
and the error bars denote 1 standard
deviation.
0.4 , . . . - - - - - - - - - - - - - . . ,
• 15 Daphnia magla
0 Daptwia magla
D
02
,-
<5
CD
~
a
J:::
3o
(;
--JIfl-----~-----~-
-02
-0.4
With macrophytes
Without macrophytes
Mid-lake
heterotrophic nanoflagellates needs to be fairly high if they are to survive as a
population during intensive Daphnia growth.
A laboratory experiment conducted with D. magna and HNF populations from
Ring S!2i indicated that the predation rate by 15 D. magna/L roughly equaled the
growth rate of the natural HNF community as measured under predator-free
conditions during 24-hour incubation (Fig. 17.3), even though the HNF population
divided every 8 hours. This is, however, not a realistic picture of the instantaneous
Daphnia predation rates because the food concentration (i.e., the HNF abundance)
was not constant during the experiment. The measured predation rates were most
likely underestimated, owing to the fact that Daphnia are not able to reduce the
HNF abundance much below 1 x 102/ml (the lowest HNF abundance recorded in
the entire data set was 0.7 x 102/ml).
Growth rates of the HNF community (i.e., species of 2-10 11m in diameter)
were measured on a routine basis during this study under predator-free conditions
(Carrick et ai., 1992; Hansen and Christoffersen, 1995), implying that the water
had been prescreened through 20-llm nets to ensure that no large predators were
present. The measured growth rates ranged from -0.70 to 3.72/day, with an
average value of 0.87/day (SD = 0.85). A few results (10 of 123 growth experiments) had to be excluded from further analysis because the growth rates were
negative. Such results are possible, for example, if predators passed the 20-llm
mesh net or if the enclosed HNF community was somehow damaged. Linear
correlation analysis of the whole data set (base e transfonned data except for
temperature) revealed that HNF growth rates were neither correlated with temperature nor with bacterial abundance (data not shown). A correlation between
temperature and HNF growth rates covering most of the growth season in one of
the lakes (Frederiksborg Slotss!2i) was, however, significant (N = 23; = 0.410;
P <.05). A calculated QIO value of 2.3 seemed reasonable compared with previously published results (Weisse, 1991).
HNF abundance and individual biovolume were not correlated with the abundance of bacteria (Table 17.2). Previous analyses including a broader range of
lakes have shown significant (but often weak) correlations (Berninger et al., 1991;
r
280
K. Christoffersen
Sanders et al., 1992; Gasol and Vaque, 1993; Gasol, 1994). It is generally accepted
that the relationship is weakened in more eutrophic systems, because various
organisms feeding on both bacteria and HNF (ciliates, rotifers, and metazoans)
often increase in number along a trophic gradient (Sanders et al., 1992; Riemann
and Christoffersen, 1993). It was apparent from this data set that Daphnia alone
had a significant negative impact on the abundance of bacteria (Table 17.2).
Most of the HNF communities examined during this study increased dramatically in abundance when released from Daphnia predation (although alternative
predators, such as ciliates, were able to prevent HNF growth in some cases), and
HNF growth rates were not related to the bacterial abundance. This suggests that
the HNF populations in eutrophic lakes are seldom limited by food and that they
are capable of dividing several times daily on the available food.
Gasol (1994) presented a framework for the assessment of bottom-up and
top-down regulation of HNF abundance. His concept is based on simple models
describing the maximum attainable HNF abundance and mean realized abundance
in a log-log plot of bacterial abundance versus HNF abundance. The model can,
under a set of assumptions, test the relative importance of top-down or bottom-up
control of HNF abundance. It was suggested that the degree of uncoupling between bacteria and HNF should be related to Daphnia abundance. The lack of
correlation between bacteria and HNF abundance and the highly significant correlation between Daphnia and HNF abundance found in this study seem to follow
the same trends as found by Gasol (1994). The effects of Daphnia predation may
be more pronounced if the relationships were based on biomass instead of abundance because large and small cladocerans differ in feeding behavior (Jiirgens,
1994).
From the present study, which was based on a large number of observations of
bacteria, HNF, and Daphnia abundances, it was concluded that HNF populations
in shallow and eutrophic lakes have the potential to proliferate when Daphnia
abundance is below a few individuals per liter. This occurs typically at low
temperatures (spring), low macrophyte densities, and/or high fish predation. Generally, the presence of Daphnia was the most important factor for regulating HNF
population dynamics, whereas temperature and bacterial abundance were only
marginally important. It appears that the mechanisms operating directly on the
abundance and growth of Daphnia in eutrophic systems also have a strong but
indirect effect on the microbial community. This does not necessarily mean that a
one-way top-down effect is prevailing, as other factors than predation may regulate the
development of Daphnia (e.g., food limitation or toxic cyanobacterial blooms).
Acknowledgments. I am grateful to Josep Gasol, Klaus JUrgens, and Brian Moss
for valuable comments on the manuscript. Erik Jeppesen, Martin Sj/lndergaard,
Morten Sj/lndergaard, Jon T. Nielsen, and Uffe K. Rasmussen have kindly provided
data on densities of macrophytes, zooplankton, and bacteria. Nils Willumsen has
provided valuable technical assistance, and Anne Mette Poulsen made linguistic
corrections. Financial support was in part provided by the Strategic Environmental
Research Programme-Centre for Freshwater.
17. Abundance, Size, and Growth of N anoflagellates
281
References
Arndt, H. Rotifers as predators on components of the microbial web (bacteria, heterotrophic
flagellates, ciliates)-a review. Hydrobiologia 2551256:231-246; 1993.
Arndt, H. Protozoen als wesentliche Komponente pelagischer Okosysteme von Seen.
Kataloge Landesmuseum N. F. 71:111-147; 1994.
Berninger, U-G.; Finlay, B.J.; Kuuppo-Leinikki, P. Protozoan control of bacterial abundance in freshwater. Limno1. Oceanogr. 36: 139-147; 1991.
Carrick, H.J.; Fahnenstiel, GL; Taylor, W.D. Growth and production of planktonic protozoa in Lake Michigan: in situ versus in vitro comparisons and importance to food web
dynamics. Limnol. Oceanogr. 37:1221-1235; 1992.
Christoffersen, K.; Riemann, B.; Klysner, A.; S~ndergaard, Mo. Potential role of fish
predation and natural populations of zooplankton in structuring a plankton community
in eutrophic lake water. Limnol. Oceanogr. 35:1429-1436; 1993.
Gasol, J.M. A framework for the assessment of top-down vs. bottom-up control of heterotrophic nanoflagellate abundance. Mar. Ecol. Prog. Ser. 113:291-300; 1994.
Gasol, J.M.; Vaque, D. Lack of coupling between heterotrophic nanoflagellates and bacteria: A
general phenomenon across aquatic systems? Limnol. Oceanogr. 38:657-665; 1993.
Giide, H. Direct and indirect influences of crustacean zooplankton on bacterioplankton of
Lake Constance. Hydrobiologia 159:63-73; 1988.
Hansen, B.; Christoffersen, K. Specific growth rates of heterotrophic plankton organisms in
a eutrophic lake during a spring bloom. J. Plankton Res. 17:413-430; 1995.
Jeppesen, E.; Sortkjrer, 0.; S~ndergaard, Ma.; Erlandsen, M. Impact of a trophic cascade on
heterotrophic bacterioplankton production in two shallow fish-manipulated lakes. Arch.
Hydrobio1. Beih. Ergebn. Limno1. 37:219-231; 1992.
Jeppesen, E.; S~ndergaard, Ma.; S~ndergaard, Mo.; Christoffersen, K.; Jiirgens, K.; TheilNielsen, J.; Schliiter, L. Cascading trophic interactions in the littoral zone of a shallow
lake (submitted).
Jiirgens, K. Impact of Daphnia on planktonic microbial food webs-a review. Mar. Microb.
Food Webs 8:295-324; 1994.
Jiirgens, K.; Stolpe, G. Seasonal dynamics of crustacean zooplankton, heterotrophic nanoflagellates and bacteria in a shallow, eutrophic lake. Freshwat. BioI. 33:27-38; 1995.
Pace, ML; Funke, E. Regulation of planktonic microbial communities by nutrients and
herbivores. Ecology 72:904--914; 1991.
Riemann, B. Potential importance of fish predation and zooplankton grazing on natural
populations offreshwaterbacteria. Appl. Environ. Microbiol. 50:187-193; 1985.
Riemann, B.; Christoffersen, K. Microbial trophodynamics in temperate lakes. Mar. Microb.
Food Webs 7:69-100; 1993.
Sanders, R.w.; Caron, D.A.; Berninger, U-G. Relationships between bacteria and heterotrophic nanoplankton in marine and fresh waters: an inter-ecosystem comparison. Mar.
Eco1. Prog. Ser. 86:1-14; 1992.
Sanders, R.W.; Porter, K.G. Bacteriovorous flagellates as food resources for the freshwater
crustacean zooplankton Daphnia ambigua. Limnol. Oceanogr. 34:673-687; 1990.
Sanders, R.w.; Leeper, D.A.; King, C.H.; Porter, K.G. Grazing by rotifers and crustacean
zooplankton on nanoplanktonic protists. Hydrobiologia 288:167--181; 1994.
Schriver, P.; B~gestrand, J.; Jeppesen, E.; S~ndergaard, Mo. Impact of submerged macrophytes on the interactions between fish, zooplankton and phytoplankton: large-scale
enclosure experiments in a shallow lake. Freshwat. BioI. 33:255-270; 1995.
Stoeckner, D.K.; Capuzzo, J.M. Predation on protozoa: its importance to zooplankton. J.
Plankton Res. 12:891-908; 1990.
Vaque, D.; Pace, ML Grazing on bacteria by flagellates and cladocerans in lakes of
contrasting food-web structure. J. Plankton Res. 14:307-321; 1992.
Weisse, T. The annual cycle of heterotrophic freshwater nanoflagellates: role of bottom-up
vs. top-down control. J. Plankton Res. 13: 167-185: 1991.
18. What Do Herbivore Exclusion Experiments Tell
Us? An Investigation Using Black Swans (Cygnus
atratus) and Filamentous Algae in a Shallow Lake
Robert T. Wass and Stuart F. Mitchell
Introduction
One common approach to the problems of quantifying herbivory is to determine
how much plant material herbivores eat. Another is to exclude the herbivores
experimentally and compare the performances of the grazed and ungrazed plant
communities. Both approaches may give biased estimates of the interaction
strength or effect of grazers on plant biomass increase or productivity (Mitchell
and Wass, 1996a). Simple consideration of the fraction of annual plant productivity
consumed ignores the effect of the time at which the material is consumed. Because
grazing removes not only biomass but also the future productive potential of that
biomass, consumption of small amounts of plant tissue early in a plant growth cycle
has a greater effect than similar amounts of consumption later (e.g., Kiprboe, 1980). It
also neglects the indirect feedback effects of herbivores, such as nutrient recycling,
damage to the plants, or relief of density suppression of growth (e.g., Lodge, 1991).
Potential bias in herbivore exclusion experiments arises largely from the particular experimental period chosen. The sensitivity of plants to early grazing
makes the starting time and conditions critically important. Earlier onset of density
limitation in the ungrazed plants than in the grazed plants may make the time
chosen to end the experiment equally important for its apparent outcome and may
provide spurious indications of stimulation of productivity by grazing. Finally,
results of such experiments have often been quantified with formulations that are
static or otherwise inappropriate (Mitchell and Wass, 1996a).
282
18. Herbivore Exclusion Experiments
283
Differences between plant biomasses in grazed and ungrazed plots are still
often considered to represent grazing consumption or removal (e.g., Cyr and Pace,
1993; Cattaneo and Mousseau, 1995). We have argued that they do not (Mitchell
and Wass, 1996a). In addition to the confounding effects of timing and density
dependence, ungrazed plants do not experience the feedback effects of herbivores,
which are part of the natural system. There is also an effect (biomass compounding) from the biomass dependence of productivity, which can be expected to cause
the differences between the biomasses to overestimate consumption or removal
by amounts that increase exponentially until density limitation is reached. This
artifact can be removed by expressing biomass changes and grazing consumption
as tissue-specific rates (e.g., Geertz-Hansen et aI., 1993).
Concurrent use of the two approaches, in combination with a simple exponential model of plant growth, provides a potential means of isolating the grazing and
net feedback effects of herbivores on plant biomass increase or productivity
(Mitchell and Wass, 1996a). A common formulation is
Net productivity = B (Big + C + L)
(1)
where B is the initial plant biomass, and Big, C, and L are, respectively, the
instantaneous rates of biomass increase in the grazed plants, grazing consumption,
and losses due to decay, organic secretion, and export. This or equivalent formulations have been widely used in studies of grazing and plant production. It is
unbiased. An alternative equation is
Net productivity = B (Biu + L)
(2)
where Biu is the instantaneous rate of ungrazed biomass increase. This equation is
biased by the absence of feedback effects of the herbivores on the plants. Enumeration
of the terms in the two equations may allow the net feedback effect to be quantified.
Our objectives were to test this model in a system in which feedback effects
were likely to be negligible, so that the two equations become essentially equivalent and to evaluate the use of the two approaches for estimating grazing consumption and interaction strength. The investigation formed a part of wider studies
on the role of black swans in shallow lake ecosystems. The study was carried out
at Hawksbury Lagoon, New Zealand, where black swans are the only large
deep-feeding herbivores, in 1990-1991, when the benthic vegetation consisted
almost entirely of filamentous green algae. Absence or near-absence of feedback
was inferred from the negligible contribution of swan excreta to nutrients in the
system (Mitchell and Wass, 1995) and the lack of morphological differentiation in
the algae, which eliminates the potential effects of damage to individual whole
plants, selective grazing, and resource allocation by the plants. Any algae detached
by the swans and removed from the control area by water currents were expected
to be largely replaced by a similar influx from the surrounding waters. Another
potential feedback, bioturbation by swans, appears to be insignificant for the
benthic light climate (Mitchell and Wass, 1996b).
284
R. T. Wass and S.F. Mitchell
=
The main basin of Hawksbury Lagoon is shallow (z 0.4-0.6 m), flat-bottomed, highly eutrophic, and brackish. It has an area of 25 ha and is located at 46°
S in South Island, New Zealand. It shows irregular nonseasonal cycles of dominance by phytoplankton and by benthic algae. These cycles are closely reflected
by changes in the black swan population, with birds migrating to or from the lake
in response to both the long-term changes in benthic algal biomass and often also
the week-to-week changes (Mitchell and Wass, 1995, 1996b). Swans show no diel
migration between the lake and the surrounding land. They feed actively at night,
and this population consumes little terrestrial food (McKinnon, 1989). Predation
and breeding recruitment at the lake itself are insignificant.
Materials and Methods
A single rectangular swan exclosure, 20 x 20 m, and a similar neighboring control
area were established near the center of the lake, in a region that appeared to be
fairly representative of the lake in terms of both swan use and benthic algal
biomass. Long-term studies have shown that swan occupancy of the region of the
study area was similar to that in the whole lake. The control and experimental
areas were defined by wooden stakes. Plastic mesh 0.5 m high was nailed to the
stakes above the water line of the exclosure, to exclude swans but to avoid
impeding the exchange of water with the lake. Duplicate algal samples were taken
randomly from each of 21 sample blocks within each area by wading along
defined Walkways. Samples were taken with a 0.004-m2 core sampler, collected by
sieving on a 1-mm mesh, washed, dried, and weighed (McKinnon and Mitchell,
1994). Swans were counted from the shore.
Exponential rates of biomass increase (per day [i.e., gIg/day]) were calculated
for the grazed and ungrazed algae for each sampling interval as (In BI - In Bo)lt,
where Bo and BI were the initial and final arithmetic mean biomasses and t is time
in days. Tissue-specific grazing rates (per day) were similarly calculated from
plant biomass and grazing consumption, from the feeding rate of 105 g DWIswan!
day. This figure was derived near the end of our study by using cellulose as an
indigestible food marker, with fecal production being determined by two independent methods (Mitchell and Wass, 1995). Instantaneous loss rates were not
determined but were assumed to be density-independent.
Results
Algal biomass increased rapidly in the exclosure from the beginning of the
experiment in the southern autumn of 1990 (Fig. 18.1). The increase was highly
significant (t-test, In transformed biomass, P <.01) but brief. Control biomass
remained unchanged through this period. The swan popUlation was fairly high
through winter (Fig. 18.1) but had no detectable effect on the algal biomass. A
grazing effect quickly became apparent with the onset of spring growth. The
18. Herbivore Exclusion Experiments
200
N
e
---G-
~
160
'"0
S
:.0
120
'"'"
<ii
OJ)
--+--
285
Grazed
Ungrazed
80
:;:
.~
oSc:
40
Q.)
a:l
0
t.s
14
12
os:: 10
§'"
;t
CI)
8
6
4
2
Figure 18.1. Benthic algal biomass (dry weight) in the enclosure and control areas of
Hawksbury Lagoon and swan population density 1990-1991. Error bars = SE.
ungrazed biomass increased at a rate, fitted by regression, of O.042/day (r = 0.99,
95% confidence interval = 0.036-0.048/day) until November, when the algae
apparently became density limited. By this time, they had grown to the surface,
forming a thick mat across the exclosure, and there was no further significant
increase in biomass. The biomass of the grazed community continued to increase,
in close accordance with the exponential model, at a rate of 0.0 17/day until the end
of the study in March (r = 0.95, 95% confidence interval = 0.012-0.0211day). By
March, the ungrazed and grazed biomasses were no longer significantly different,
so that the net effect of grazing in the spring-autumn period was effectively zero.
Algae identified during the study were Enteromorpha intestinalis, Rhizoclonium
sp., and Spirogira sp. Trace amounts of Ruppia megacarpa and Nitella hookeri
were recorded occasionally. The algal biomass was 2 orders of inagnitude higher
at the end of the experiment than at the beginning, a year earlier. Intermittent
observations indicated that it remained high for the next 2 years, before collapsing
in autumn 1993 (Mitchell and Wass, 1996b).
The difference between the growth rates of the control and experimental algae
implies an average spring grazing rate of 0.025/day. The rates were, however,
variable. For example, the direct estimates were as high as 0.042/day in JulyAugust but declined progressively as the swan population density failed to keep
pace with the increasing algal biomass (Table 18.1; Fig. 18.1). By the end of the
study they were less than O.OOI/day.
3.1- 5.6 (0.37)
5.6-10.1 (0.73)
10.1- 9.8 (1.10)
9.8-18.1 (1.51)
Sampling
interval 1990
Aug 19-5ept 10
Sept 10-0ct 1
Oct I-Oct 22
Oct 22-Nov 12
0.027
0.028
-0.001
0.029
Grazed
biomass
increase (Big)
(per day)
0.022
0.Ql5
0.009
0.008
Mean grazing
rate (C)
(per day)
2.3- 7.8 (0.32)
7.6-16.5 (1.05)
16.5-38.9 (3.1)
38.9-84.6 (2.97)
Initial and final
ungrazed
biomass (glm 2)
(SE initial
biomass, n = 21)
0.055
0.039
0.041
0.037
Ungrazed
biomass
increase (Biu)
(per day)
0.152
0.241
0.081
0.363
Production
+ C)
(g m-2/day)
B(Big
GLoss rates are not accounted for. Grazing rates are geometric means of initial and final direct estimates for the sampling interval. Biomass is DW.
Initial (B) and
fi nal grazed
biomass (glm 2)
(SE initial
biomass, n = 21)
Table IS. 1. Net Benthic Algal Productivity in Hawksbury Lagoon, Calculated from Equations 1 and 2a
0.170
0.218
0.414
0.363
BxBiu
(g m-2/day)
Production
287
18. Herbivore Exclusion Experiments
3~---------------------------------------,
~
<;'
T
2.5
E
2
~
1.5
-I+-----~----~----~----~----r_----r_--~
19 Aug
10 Sep
1 Oct
22 Oct
12 Nov
3 Dec
12 Jan
10 Mar
Sample period ending
Figure 18.2. Net productivity of filamentous algae in Hawksbury Lagoon in spring-suMmer 1990-1991.. = productivity of ungrazed algae calculated from biomass changes;O=
productivity of grazed algae, calculated from biomass changes ( 0 ) plus amounts removed
by swan grazing. Error bars = SE for biomass change.
.
Results of the two methods of calculating productivity agreed closely in three
of the four sampling intervals for which they could be compared (Table 18.1).
There was, however, a large difference in the interval from October 1-22, when
the biomass of the grazed algae declined. We were unable to test this model after
mid-November, as the ungrazed algae had become density limited. The productivity of the grazed algae was higher than that of the ungrazed community in two
of the three subsequent samples (Fig. 18.2).
Annual net productivity (April-March), calculated from grazed biomass increases plus grazing consumption, was 185 g DW/m2, of which 22 g DW/m2 or
12% was removed by grazing.
Discussion
The agreement between the two independent productivity estimates in three of the
four tests indicates that our approach may be useful for isolating direct consumption and feedback effects in other systems. The discrepancy in the other test might
suggest that there was a large negative feedback in this interval. As feedback
effects can be expected to act continuously, however, it seems more likely to
represent a violation of the assumption that loss rates were density-independent.
The grazed and ungrazed algae might, for example, differ in their susceptibility to
being dislodged by waves. Also, the ungrazed algae that had grown toward the
lake surface must have experienced a different light climate from the natural
community, confined by grazing to the bottom few centimeters. Although the
water was generally clear, this effect was potentially important during any resuspension events, as rates of decline in algal biomass in the lake during periods of
288
R.T. Wass and S.F. Mitchell
low benthic light may be high (Mitchell and Wass, 1996b ). Such effects will
obviously become increasingly likely as the biomasses diverge. It is highly desirable that loss rates should be quantified in any further attempts to use this model.
Alternatively, the effect of any density dependence of loss rates could be reduced
by conducting sequential short-term experiments throughout a growth cycle. It is
also clear that ability to detect and estimate feedback will be subject to the high
sampling variation in most plant communities. At present very little is known of
the magnitude of feedback effects, although Lodge (1991) among others has
argued that they are potentially very important.
The final outcome of the experiment was that swans had no significant effect on
the algal biomass. If we had ended the experiment earlier or begun it later, we
might have reached a different conclusion. For example, on November 12 the
grazed biomass was only 17% of the ungrazed biomass (Fig. 18.1). It is also
predictable that if we had delayed the start of the experiment, the seasonal
dynamics would have been different, with the ungrazed algae reaching the carrying capacity progressively later as the delay increased. These results reinforce the
doubts that we have expressed about the interpretation of herbivore exclusion
experiments conducted over arbitrary experimental periods, without reference to
plant growth cycles (Mitchell and Wass, 1996a).
Our results also illustrate the reality of biomass compounding. Algal productivity estimates, uncorrected for this effect, are shown in Figure 18.2 with the
estimates from equation 1. At the extreme (October 22-November 12), the difference between the ungrazed and grazed biomasses indicated a consumption or
removal of 83% of the net primary productivity or biomass, whereas the correctly
formulated estimate for this interval was 14%, and cumulative consumption from
August 19 to November 12 was only 10% of the final ungrazed biomass. Similar
differences between direct estimates of algal consumption and indirect estimates
based on the crude difference between grazed and ungrazed biomasses are common in periphyton communities (Cattaneo and Mousseau, 1995). These authors
attributed them to herbivore-induced losses of algae by sloughing. The level of
agreement between the estimates from equations 1 and 2 confirms that this was not
so in our study. Herbivores can neither remove nor eat material that has never been
present in the real (grazed) system.
The higher algal productivity in the control area than in the enclosure in the last
two sample intervals was not due to stimulation of productivity by the herbivores.
It occurred simply because the ungrazed algae had become density limited and the
grazed algae had not.
We conclude that grazing consumption cannot normally be determined without
studying grazers. It can be estimated from herbivore exclusion experiments only by
use of tissue-specific rates, when there are no feedback effects of the herbivores on the
plants, and while plant growth remains density-independent. Interaction strengths can
be understood and quantified adequately only by dynamic analysis, preferably over
complete natural cycles of plant biomass increase and decline.
A potential increase in the algae was completely suppressed by swan grazing in
autumn 1990. Opportunity for this effect to occur is, however, very limited. At the
18. Herbivore Exclusion Experiments
289
average spring growth rate ofO.042/day, an algal biomass of only 2 glm2 would be
required for production to meet the food requirements of the 8-10 swans/ha that
were present. Black swans can also play little role in the loss of benthic algae when
the lake switches to phytoplankton dominance, as the final biomass represents
several years' consumption by even the highest swan population ever observed
there.
The high benthic algal biomass in autumn 1991 and the subsequent 2 years of
benthic algal dominance of the lake are in contrast to 1994, when turbidity from
phytoplankton and resuspended sediment caused Nitella to collapse in FebruaryMarch (Mitchell and Wass, 1996b). Black swans contributed little to either outcome.
Acknowledgments. The study was supported by a University of Otago Research
Grant. We are also grateful to Brian Niven and Brian Manly of the Department of
Mathematics and Statistics at Otago University for statistical advice.
References
Cattaneo, A.; Mousseau, B. Empirical analysis ofthe removal rate of periphyton by grazers.
Oecologia. 103:249-254; 1995.
Cyr, H.; Pace, M.L. Magnitude and patterns of herbivory in aquatic and terrestrial ecosystems. Nature 361:148-150; 1993.
Geertz-Hansen, 0.; Sand-Jensen, K.; Hansen, D.E; Christiansen, A. Growth and grazing
control of the abundance of the marine macroalga Ulva lactuca L. in a eutrophic Danish
estuary. Aquat. Bot. 46:101-109; 1993.
Kij2jrboe, T. Distribution and production of submerged macrophytes in Tipper Grund (Ringkj2jbing Fjord, Denmark), and the impact of waterfowl grazing. J. App!. Eco!. 17:675687; 1980.
Lodge, D.M. Herbivory on freshwater macrophytes. Aquat. Bot. 41: 195-224; 1991.
McKinnon, S.L. The interrelationship between phytoplankton, submerged macrophytes and
black swans (Cygnus atratus) in New Zealand lakes-test of two models. MSc thesis,
University of Otago, Dunedin; 1989.
McKinnon, S.L.; Mitchell, S.P. Eutrophication and black swan (Cygnus atratus Latham)
populations: tests of two simple relationships. Hydrobiologia 279/280: 163-170; 1994.
Mitchell, S.P.; Wass, R.T. Food consumption and faecal deposition of plant nutrients by
black swans (Cygnus atratus Latham) in a shallow New Zealand lake. Hydrobiologia
306:189-197; 1995.
Mitchell, S.P.; Wass, R.T. Quantifying herbivory-grazing consumption and interaction
strength. Oikos 76:573-576; 1996a.
Mitchell, S.P.; Wass, R.T. Grazing by black swans (Cygnus atratus Latham), physical
factors, and the growth and loss of aquatic vegetation in a shallow lake. Aquat. Bot.
55:205-215; 1996b.
19. Switches Between Clear and Thrbid Water
States in a Biomanipulated Lake (1986--1996):
The Role of Herbivory on Macrophytes
Ellen Van Donk
Introduction
Shallow lakes may display alternate stable states over a range of nutrient concentrations, a clearwater state dominated by aquatic vegetation, and a turbid state
characterized by high algal biomass (Scheffer et aI., 1993). This may have important implications for the possibilities of restoring eutrophied shallow lakes.
Man-made modification of fish populations ("biomanipulation") has been applied successfully to severnl small shallow lakes to induce a transition from a
phytoplankton-dominated state to a clearwater state with submerged macrophytes
(e.g., Hanson and Butler, 1994; Lauridsen et al., 1994b; Meijeret aI., 1994). These
macrophytes playa key role in severnl mechanisms that tend to keep the system in
a clearwater state at relatively high nutrient loadings (Jeppesen et aI., this volume,
Chapter 5). An important question underlying the use of biomanipulation as a
restoration technique is its long-term effectiveness at different nutrient loadings
(Jeppesen et al., 1990). Most studies published thus far lasted less than 5 years,
whereas according to Frost et al. (1988), studies of fish manipulations, which
involve complex interactions, should extend for longer periods.
In Lake Zwernlust, a eutrophic lake located in the middle of The Netherlands,
biomanipulation measures (major changes in total fish community) were taken in
March 1987 (van Donk et al., 1989). After this biomanipulation, the lake shifted
several times between the turbid and the clearwater state. In this chapter, I give an
overview ofthis switching behavior over a lO-year period (1986-1996) in relation
290
19. Water States in a Biomanipu1ated Lake
291
to changes in macrophyte composition and herbivory by waterfowl (coots) and
fish (rudd). An account of the period until 1994, characterized by a return of
turbidity after a clear period of several years, was given previously (van Donk and
Gulati, 1995). The present extension until 1996 covers an autonomous but temporary recovery of a clearwater state.
Methods
Lake Zwemlust is a small water body (area, 1.5 ha; mean depth, 1.5 m) situated in
the middle of The Netherlands. The external P and N loadings to the lake are high
and estimated at approximately 2 g P/rnJyr and 9 g NlrnJyr, respectively. Before
biomanipulation in 1987, phytoplankton blooms were dominant during the whole
year (Secchi depth, 0.3 m). Extensive description of the limnology of the lake,
before and after biomanipulation, is given in van Donk et al. (1990, 1993) and van
Donk and Gulati (1995). The biomanipulation measures are discussed at some
length by van Donk et al. (1989, 1990).
Biomass and composition of submerged macrophytes in the lake were estimated according to Ozimek et al. (1990). The method of Prejs (1984) was
followed to estimate the consumption of macrophytes by rudd. As for coot grazing
on submerged macrophytes, a daily intake of about 45 g DW was found by Hurter
(1979). The consumption of macrophytes by coots was estimated from this daily
consumption per coot and the number of "birds days" (average number of birds
per day multiplied by number of days).
Exclosures were used to evaluate grazing effects by fish and birds on macrophyte species composition. Cages made of an iron frame with dimensions of 4 m
(length) x 1.5 m (width) x 0.6 m (height) and covered by wire netting served as
exclosures for larger fish and birds. These were placed on the lake bottom at a
depth of 2.0 m in May 1992, and the experiment lasted until July 1993. Initially,
macrophyte characteristics (species composition and biomass) in the cages were
similar to those in the lake. At the end of this experimental period, the percentages
of vegetated area occupied by the different macrophyte species inside and outside
the cages were determined. The design of these experiments is described in more
detail in van Donk and Otte (1996).
Results and Discussion
Switches Between the Different States Related to Macrophyte Species
Composition (1986-1996)
After application of biomanipulation in 1987, the lake shifted within 1 year from a
turbid state dominated by phytoplankton and no submerged vegetation (state I in
Fig. 19.1 and Table 19.1) to a clearwater state dominated by submerged macrophytes (state III). From 1988 until 1992, the lake remained clear during the whole
292
E. Van Donk
_
100
90
~ Birds (coots)
Fish (rudd)
III
I
III
II
C.d.
P.b.
~
->.
80
1:
70
0
60
~
~
c:
aE
0
E.n.
II
P.b.
50
40
::J
Cfl
30
0
()
20
c:
NoVeg.
III
10
0
1~
lE1~
1~
1m 1m
1~
1~
1~1~
1~
l' biomanipulation
Figure 19.1. Estimates of annual macrophyte consumption by fish (rudd) and birds (coots)
(macrophyte g DW/mJyr) before and after biomanipulation in Lake Zwemlust. The three
different states (see Table 19.1) and the dominant submerged macrophyte species are given.
State I, turbid state; state II, clear/turbid state; state III, clearwater state. No Veg, no
submerged vegetation; E.n., Elodea nuttallii; C.d., Ceratophyllum demersum; P.b., Potamogeton berchtoldii.
year, mainly due to macrophyte-induced nitrogen limitation of the phytoplankton
growth during summer and autumn (van Donk et aI., 1993; van Donk and Gulati,
1995). In the first 2 years after biomanipulation (1988 and 1989), Elodea nuttallii
was the dominant macrophyte species. Elodea is a perennial submerged macrophyte and rather insensitive to lower temperatures in late autumn and winter. In the
subsequent period of 2 years (1990 and 1991), Ceratophyllum demersum became
more dominant. Both Elodea and Ceratophyllum are able to compete strongly with
phytoplankton for nutrients, especially for nitrogen (Best, 1977; van Donk et aI.,
1993).
In 1992, 1993, and 1994, the lake was turbid during late summer and autumn,
but otherwise remained clear. In these years, C. demersum and E. nuttallii were
nearly absent and Potamogeton berchtoldii became the dominant species in spring,
declining to very low abundance during late summer when it started to form
turions (overwintering structures). Potamogeton species are known to form these
structures in autumn (Sastroutomo, 1981). P. berchtoldii was progressively
covered with epiphytes from late spring to summer (van Donk and Gulati, 1995).
No significant growth of epiphytes was observed on the other macrophyte species.
The early collapse of Potamogeton was probably caused by light limitation due to
epiphyte growth, inducing early formation of turions (van Donk and Gulati, 1995).
Relative abundance of phytoplankton increased after the collapse of macrophytes
and the rise of epiphytes. Consequently, the lake shifted to a turbid state (state II)
Perennials (e.g., Elodea)
Herbiv
Pisciv
Planktiv
Benthiv
Herbiv
Pisciv
Fish
<30 whole year
<30 (spring/summer)
30-80 (autumn)
80-250 whole year
Phytoplankton
chlorophyll a (llgIL)
"Planktiv, planktivorous; benthiv, benthivorous; herbiv, herbivorous; pisciv, piscivorous.
III
li;learL
No
I
(Turbid)
II
(Clear/turbid)
Species forming
overwintering
structures
(e.g., Potamogeton)
Submerged macrophytes
State
Large
daphnids
Daphnids
Rotifers
Rotifers
Zooplankton
Light limitation
Zooplankton &
N-limitation
(spring/summer)
Light limitation (autumn)
Zooplankton-grazing
N-limitation
No
Herbiv
Phytoplankton limitation
No
Birds
Table 19.1. The Three Different States and Their Ecosystem Structures as Observed in Lake Zwemlust over a 10-Year Period (1986-1996)"
294
E. Van Donk
but only during late summer and autumn. State II differs from state I in that a
clearwater state still exists during winter and early summer (Table 19.1).
In 1995, E. nuttallii reappeared and became the dominant macrophyte species.
The clearwater state (state III) was re-established during the whole year. In 1996,
however, when Elodea was very scarce and Potamogeton abounded, the lake
shifted again to the turbid state II.
Shifts in Macrophyte Species Composition Related to Herbivory
Herbivory by vertebrate and invertebrate grazers may play an important role in
structuring and lowering the macrophyte biomass (e.g., Lodge, 1991; Lodge et aI.,
this volume, Chapter 8). In Lake Zwemlust, herbivory by coots (Fulica atra) and
rudd (Scardinius erythrophthalmus) was observed after re-establishment of the
macrophytes (van Donk et al., 1994).
Extensive grazing of coots on submerged macrophytes occurs mainly during
autumn and winter, after territories break up (Kiy;rboe, 1980; Perrow et aI., 1997).
Kiy;rboe (1980) stated that grazing by coots has only a minimal effect on macrophyte growth because grazing often takes place outside the growing season of the
plants. Coots, however, pull out whole plants and may influence the macrophyte
composition and succession by removing especially plants still present during
autumn and winter. A high number of coots appeared in Lake Zwemlust when
perennial macrophytes became dominant after biomanipulation (Fig. 19.1). An
increase of herbivorous birds, after the restoration of submerged macrophyte
communities, was also observed in various other lakes (Hanson and Butler, 1994;
Hargeby et aI., 1994; Lauridsen et al., 1994a; Schutten et aI., 1994; Sy;ndergaard et
al., this volume, Chapter 20). High grazing pressure on Elodea and Ceratophyllum
during the fIrst years after biomanipulation probably promoted the rise of Potamogeton (van Donk and Gulati, 1995). P. berchtoldii was not negatively affected by coots
because this species forms nongrazable overwintering structures. From 1992 to 1995,
virtually no coots were present, probably due to absence of submerged vegetation
during autumn and winter. After Elodea became abundant again in 1995, coots
returned in winter 1995/1996 and started to graze on Elodea. In 1996, Potamogeton
reappeared, followed by phytoplankton blooms during the summer.
Rudd was introduced in Lake Zwemlust in 1987 as food for pike (van Donk et
aI., 1989). According to Prejs (1984), only larger rudd (>1 +) are herbivorous. The
number of larger rudd was quite low until 1990 but increased ip 1991 to 297 kg/ha
and stabilized in 1992-1996 at 200-300 kg/ha. Rudd only graze during the
growing season of the macrophytes (temp, >16°C) and had therefore probably a
much lower impact on total macrophyte biomass compared with coots (van Donk
et aI., 1994) (Fig. 19.1). Prejs (1984) even suggested that grazing by rudd may
stimulate the growth of the plants. In laboratory experiments, rudd were observed
to consume mainly Elodea and Potamogeton but not Ceratophyllum, which is
calcareous in structure and has apparently a much lower edibility (Prejs and
lackowska, 1978; Teule, 1993). In this way, rudd may, like coots, induce a change
in macrophyte species composition.
19. Water States in a Biomanipulated Lake
295
In the cages designed to exclude herbivory by both larger fish and birds, P.
berchtoldii was the dominant species at the start of the experiment in spring 1992.
However, in early spring 1993, the exclosures became dominated by E. nuttallii,
whereas in the lake, P. berchtoldii was still abundant. The coverage percentage of
macrophytes in the exclosures was higher than that in the lake (van Donk and Dtte,
1996). In cages excluding grazing by coots, Sl'indergaard et al. (1996) found a
significantly higher macrophyte biomass than outside the exclosures, but in contrast to our findings, no shift in macrophyte species composition was reported.
Further results of the exclosure experiments are described in van Donk and Dtte
(1996).
Conclusions
Ten years after application of biomanipulation, Lake Zwemlust has undergone
several switches between turbid and clearwater states. The presence of submerged
macrophytes seems to be essential in keeping this lake clear. The observations
suggest that dominance of perennials such as Elodea and Ceratophyllum leads to
a whole-year clearwater state, whereas species forming overwintering structures,
such as Potamogeton, give rise to epiphyte growth in early summer and phytoplankton blooms in late summer and autumn. Herbivory, especially by coots, was
probably an important factor in triggering the shift from perennials to nonperennials. The switch in 1995 from state II (turbid) back to state III (clear) was rather
unexpected. Probably, the perennial Elodea was able to re-establish itself due to
absence of coots in the previous winters. This conclusion is based on the results of
the exclosure experiments. Exclosures, however, can lead to misleading conclusions because they may also change other factors such as light, climate, and
water currents. Rl'irslett et al. (1986) observed that Elodea canadensis suddenly
collapsed after some years of growth and then returned to its previous dominance
years after, without any known external factors. In follow-up studies, not only the
consumption but also the production of the macrophytes has to be considered.
Also, more specific experimental research is needed to quantify the effect of
herbivory on growth of different macrophyte species and consequently on stability
of the clearwater state.
References
Best, E.P.H. Seasonal changes in mineral and organic components of Ceratophyllum
demersum and Elodea canadensis. Aquat. Bot. 3:337-348; 1977.
Frost, T.M.; DeAngelis, D.L.; Bartell, S.M.; Hall, DJ.; Hulbert, S.H. Scale in the design and
interpretation of aquatic community research. In: Carpenter, S.R., ed. Complex interactions in lake communities. New York: Springer-Verlag, 1988:229-258.
Hanson, M.A.; Butler, M.G. Responses to food web manipulation in a shallow waterfowl
lake. Hydrobiologia 2791280:457--466; 1994.
Hargeby, A.; Anderssen, G.; Blindow, I.; Johansson, S. Trophic web structure in a shallow
eutrophic lake during a dominance shift from phytoplankton to submerged macrophytes.
Hydrobiologia 279/280:83-90; 1994.
296
E. Van Donk
Hurter, H. Nahrungsokologie des Bliisshuhn (Fulica atra) an den Uberwinterungsgewiissem im nordlichen Alpenvorland. Der Omitologische Beobachter 76:257-288; 1979.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; Sf/lndergaard, M.; Mortensen, E.; Sortkjrer, 0.;
Olrik, K. Fish manipulation as a lake restoration tool in shallow, eutrophic, temporate
lakes 2: threshold levels, long-term stability and conclusions. Hydrobiologia 2001201:
219-227; 1990.
Kif/lrboe, T. Distribution and production of submerged macrophytes in Tripper Ground, and
the impact of waterfowl grazing. J. Appl. Ecol. 17:675-687; 1980.
Lauridsen, T.L.; Jeppesen, E.; 0stergaard Andersen, E Colonization and succession of
submerged macrophytes in shallow fish manipulated Lake Vreng: impact of sediment
composition and waterfowl grazing. Aquat. Bot. 46:1-15; 1994a.
Lauridsen, T.L.; Jeppesen, E.; Sf/lndergaard, M. Colonization and succession of submerged
macrophytes in shallow Lake Vreng during the first five years following fish manipulation. Hydrobiologia 275/276:233-242; 1994b.
Lodge, D.M. Herbivory on freshwater macrophytes. Aquat. Bot. 41:195-224; 1991.
Meijer, M.-L.; Jeppesen, E.; van Donk, E.; Moss, B.; Scheffer, M.; Lammens, E.; Van Nes,
E.; Faafeng, B.A; Jensen, J.P. Long-term responses to fish-stock reduction in small
shallow lakes: interpretation of five year results of four biomanipulation cases in The
Netherlands and Denmark. Hydrobiologia 275/276:457-467; 1994.
Ozimek, T.; van Donk, E.; Gulati, RD. Can macrophytes be useful in biomanipulation of
lakes? The Lake Zwemlust example. Hydrobiologia 200/201:399-409; 1990.
Perrow, M.R; Schutten, J.; Howes, J.R; Holzer, T.; Madgwick, EJ.; Jowitt, A.J.D. Interactions between coot (Fulica atra) and submerged macrophytes: the role of birds in the
restoration process. Hydrobiologia 3421343:241-255; 1997.
Prejs, A Herbivory by temperate freshwater fishes and its consequences. Environ. BioI.
Fish. 10:281-296; 1984.
Prejs, A; Jackowska, H. Lake macrophytes as the food of roach (Rutilus rutilus) and rudd
(Scardinius erythrophthalmus L.) I. Species composition and dominance relations in the
lake and the food. Ekol. Pol. 26:429-438; 1978.
Rfilrslett, B.; Berge, D.; Johansen, S.W. Lake enrichment by submerged macrophytes: a
Norwegian whole-lake experience with Elodea canadensis. Aquat. Bot. 26:325-340;
1986.
Sastroutoma, S.S. Turion formation, dormancy and germination of curly pondweed, Potamogeton crispusL. Aquat. Bot. 10:161-173; 1981.
Scheffer, M.; Hosper, S.H.; Meijer M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evol. 8:275-279; 1993.
Schutten, J.; Van der Velden, A; Smit, H. Submerged macrophytes in the recently freshened
lake system Volkerak-Zoom (The Netherlands), 1987-1991. Hydrobiologia 2751276:
207-218; 1994.
Sfilndergaard, M.; Bruun, L.; Lauridsen, T.; Jeppesen, E.; Vindbrek Madsen, T. The impact
of grazing waterfowl on submerged macrophytes: in situ experiments in a shallow
eutrophic lake. Aquat. Bot. 53:73-84; 1996.
Teule, K. The effect of rudd (Scardinius erythrophthalmus L.) on the macrophyte biomass
and species composition in Lake Zwemlust (in Dutch). Report 3062. Wageningen:
Agricultural University; 1993.
van Donk, E.; Gulati, R.D. Transition of a lake to turbid state six years after biomanipulation: mechanisms and pathways. Wat. Sci. Techn. 32:197-206; 1995.
van Donk, E.; Otte, A Effects of grazing by fish and waterfowl on the biomass and species
composition of submerged macrophytes. Hydrobiologia 340:285-290; 1996.
van Donk, E.; Gulati, RD.; Grimm, M.P. Food-web manipulation in Lake Zwemlust:
positive and negative effects during the first two years. Hydrobiol. Bull. 23:19-34;
1989.
19. Water States in a Biomanipulated Lake
297
van Donk, E.; Grimm, M.P.; Gulati, R.D.; Klein Breteler, J.P.G. Whole-lake food-web
manipulation as a means to study community interactions in a small ecosystem. Hydrobiologia 2001201:275-291; 1990.
van Donk, E.; Gulati, R.D.; Iedema, A.; Meulemans, J.T. Macrophyte-related shifts in the
nitrogen and phosphorus contents of the different trophic levels in a biomanipulated
shallow lake. Hydrobiologia 251:19-26; 1993.
van Donk, E.; De Deckere, E.; Klein Breteler, J.P.G.; Meulemans, J.T. Herbivory by
waterfowl and fish on macrophytes in a biomanipulated lake: effects on long term
recovery. Verh. Int. Verein. Limnol. 25:2139--2143; 1994.
20. Macrophyte-Waterfowl Interactions:
Tracking a Variable Resource and the Impact of
Herbivory on Plant Growth
Martin S0ndergaard, Torben L. Lauridsen, Erik Jeppesen, and
Lise Bruun
Introduction
Submerged macrophytes are important for the maintenance of clear water in
shallow eutrophic lakes (Jeppesen et aI., 1990; Scheffer, 1990; Hargeby et aI.,
1994), and establishment of permanent macrophyte coverage is an important
aspect of the lake restoration process after a reduction of nutrient loading.
However, as macrophytes are subject to grazing by herbivores such as waterfowl (e.g., Lodge, 1991), it may be speculated whether waterfowl grazing can
delay recolonization and thereby lake recovery. The impact of waterfowl
grazing on macrophytes is poorly documented (Winfield, 1991), and few
studies have considered it important, examples being Jupp and Spence (1977),
who ascribed growth limitation of Potamogeton filiformis and Potamogeton
pectinatus to wave action and waterfowl grazing, and van Donk et aI. (1994),
who observed that intensive coot herbivory in Lake Zwemlust (up to 120
individuals/ha) affected Elodea biomass and species composition.
In this study, we document large temporal variations in macrophyte and waterfowl abundance and assess the effect of waterfowl grazing on the growth of
Potamogeton crispus in two shallow Danish lakes in the early stages of submerged
macrophyte recolonization.
298
299
20. Macrophyte-Waterfowl Interactions
Methods and Study Areas
The studies were undertaken in Lake V:eng and Lake Stigsholm; a detailed
description of the methods applied and the two study areas can be found in
Lauridsen et al. (1993, 1994) and Spndergaard et al. (1996). The main characteristics of the two lakes are summarized in Table 20.1.
Lake V:eng was biomanipulated in 1986 and 1987 by removing approximately
50% of the bream (Abrarnis brarna) and roach (Rutilus rutilus) fish stock (Spndergaard et aI., 1990). After fish removal, mean summer Secchi depth increased
from 0.7 m to more than 1.8 m (lake bottom, Table 20.1), primarily due to
increased top-down control by large daphnids (Spndergaard et aI., 1990; Table 20.1). Apart from short periods (a few weeks) with turbid water, the lake has
remained clear since 1987 (Jeppesen et aI., in press).
For the past decades, Lake Stigsholm has fluctuated between a macrophyte-rich
clearwater state and a macrophyte-poor turbid state. This has been documented by
palaeolimnological studies (B. Odgaard, unpublished observation) supported by
observations of the varying number of mute swan (Cygnus olor), which ranged
from 15 individuals/ha during the summer of 1967 to only 0.3 individualslha in
1969 (Skotte-Mpller, 1970).
Macrophyte coverage and percentage volume infested (PVI, calculated as the
product of percentage coverage and height divided by water depth) were estimated
by using 9-14 transects in each lake together covering the whole lake (see
Lauridsen et aI., 1994, for details). Waterfowl abundance (coot and mute swan)
was counted from 1991 and onward in Lake V:eng and in 1990, 1994, autumn
1995, and 1996 in Lake Stigsholm.
To study the influence of grazing by coot (Fulica atra) and mute swan, Potarnogeton crispus shoots were incubated at different locations (two locations in
Lake V:eng and seven in Lake Stigsholm). In Lake V:eng, the shoots were planted
Table 20.1. Morphometric and Chemical (Mean Summer) Characteristics of Lake V reng
and Lake Stigsholm
Area (ha)
Mean depth (m)
Max depth (m)
Hydraulic retention time (days)
Total phosphorus before 1987 (mg PIL)
Total phosphorus after 1987 (mg PIL)
Sec chi depth (m)
Secchi depth (m)
Chlorophyll a before 1987 (J.,lgIL)
Chlorophyll a 1987-1995 (J.,lglL)
aSecchi depth >1.2 m in 27-73% of the samplings.
Lake Vreng
Lake Stigsholm
15
1.2
1.8
21
0.13
0.08
0.7
1.2-1.8
80
19
21
0.8
1.2
5
0.15
1.2a
41
300
M.
S~ndergaard
et al.
in two substrate types at each location: mud and sand. For each location and
substrate type, eight pots, each with one P. crispus shoot, were placed in an
unprotected box and in a box protected by a 2-cm mesh chicken wire fence that
reached from the lake bottom to 20 cm above the lake surface. The boxes were
placed 0.9 m (Lake Vreng) and 0.5 m (Lake Stigsholm) below the lake surface.
Mean shoot length, and in Lake Stigsholm also total shoot length, number of
shoots, branches per shoot, and percentage stubble (calculated as the ratio of the
number of shoots [including lateral shoots] lacking an apex to the total number of
shoots x 100) per box, was measured on six to eight dates from May to August
(1989) in Lake Vreng and on four different dates during the period July 27 to
September 3, 1990, in Lake Stigsholm. Further information can be found in
Lauridsen et al. (1993) and S~ndergaard et al. (1996).
Results and Discussion
The results from Lake V reng illustrate that marked seasonal and interannual
variations in macrophyte coverage and species composition can take place during
the recolonization phase following biomanipulation-mediated improvement in
lake transparency (Fig. 20.1). Colonization started in the second and third year
after biomanipulation, and by the end of the fourth year (1990), macrophyte
coverage had reached 80% and PVI 54%. P. crispus was initially abundant, but
Elodea canadensis soon took over and became completely dominant. Elodea
coverage and PVI have varied widely from year to year, however. In autumn 1992
and summer 1993, Elodea had almost disappeared, but it regained high density
again in summer 1994 and 1995, whereas it decreased again from 80% to 12%
coverage in 1996.
Coot and mute swan number fluctuated with macrophyte density in the autumn
and winter; high numbers were recorded only in years with high macrophyte
density (Fig. 20.1). Maximum density occurred in winter 1991/1992, when up to
20 mute swans and 53 coots were recorded per hectare. Due to territorial behavior,
waterfowl density was relatively low in the summer nesting season, irrespective of
macrophyte abundance.
From 1989 to 1995, macrophyte coverage was relatively low in Lake Stigsholm, although Secchi depth was equivalent to water depth for long periods during
the summer (Table 20.1). Macrophytes were present each year, albeit that the
maximum coverage of all species (excepting macroalgae) never exceeded 30%.
The dominant species were Callitriche hennaphroditica and Potamogeton (pectinatus and berchtoldii) (Fig. 20.2). In most years, filamentous algae (mainly
Enteromorpha and Spirogyra) were present in large quantities for a short period
during the summer. Apart from filamentous algae, which occasionally reached the
surface, plant height rarely exceeded 10-20 cm, and PVI was therefore usually
25%. Macrophyte coverage during the summer was inversely correlated to chlorophyll a, and when chlorophyll a exceeded approximately 50 J.lg/L, macrophyte
coverage was low, thus indicating that macrophyte density was closely related to
20. Macrophyte-Waterfowl Interactions
~
~
"0
100
301
Elodea
80
Ql
tl 60
~
Ql
E
~
"0
40
20
~
<Ii
0
~
20
>
0
10
Cl
Ql
<t
~
io
0
30
20
~
10
"0
0
'-
50
:>
:§.
Ql
Potamogeton
Mute swan
Coot
.0
E 40
~
Z
30
20
10
0
1987
Figure 20.1. Development of submerged macrophytes (0 mean percentage coverage of
total lake area and e: mean PVI) in Lake Vreng. Coot (Fulica alra) and mute swan (Cygnus
olor) numbers are shown in the two lower panels.
changes in turbidity. By the end of 1995 and during 1996, macrophyte (mainly
Elodea) coverage and PVI increased markedly (Fig. 20.2). During summer 1996,
total coverage ranged between 50-70% and PVI between 18-26%. Mean summer
chlorophyll a in 1996 was only 13 IlglL. In summer 1990 (when the grazing
experiments were conducted), coot density ranged between 3-9 individualslha,
whereas mute swan density was 0.2 individualslha. Corresponding with increasing
macrophyte abundance, autumn densities of coot were high in 1995 and 1996
compared with the previous years, with the maximum density recorded being 38
individualslha in November 1996.
In the following, we give a rough estimation of waterfowl food consumption in
Lake V reng. Based on direct measurements of faecal output, Mitchell and Wass
(1995) calculated a mean food intake of 104 g DW/day for New Zealand black
swans (mean weight, 5.6 kg) occurring at a density of 10-20 individualslha. Mute
swan are larger, however, weighing about 10.0 kg (Killlrboe, 1980), and their food
consumption is therefore presumably higher, at about 200 g DW/day. For coot,
daily food intake has been estimated to be 45 g DW/day (Hurter, 1979). Applying
these figures to Lake V reng waterfowl densities yields a maximum food consump-
302
M. S0ndergaard et al.
100
Total
80
60
40
20
~
!:....
-0
CD
U;
CD
:§
CD
E
0
40
20
(5
0
40
CD-
20
::l
.
>
Filamentous algae
Ol
~
CD
0
40
0
20
>
0
0
0
Potamogeton
Elodea
60
40
20
ro
.c
:>
-0
c
0
5
4
3
2
no data
CD
0
40
E
::l
20
.0
Z
Mute swan
Coot
no data
0
250
Chlorophyll a
200
~ 150
::::I..
~
100
50
0
Figure 20.2. Development of submerged macrophyte coverage (0 mean percentage of
total lake area and .: PVI), chlorophyll a, and waterfowl abundance in Lake Stigsholm
during the period 1989-1995.
20. Macrophyte-Waterfowl Interactions
303
tion of 4 kg DW plant materiaVday/ha for mute swan and 2.4 kg DW/day/ha for
coot. For the whole winter period of November 1, 1991, to March 1, 1992 (from
which we have the most comprehensive waterfowl data and when density was
highest), total waterfowl consumption is estimated to be 440 kg DW/ha. By
comparison, estimated maximum Elodea biomass in 1990 was 1,660 kg OW/ha
(Lauridsen et aI., 1994). Thus, assuming no net production by Elodea during the
winter, the above estimates indicate that at maximum macrophyte density (e.g., as
in 1990 and 1991 in Lake Vreng), waterfowl grazing removes only about 25% of
the macrophyte biomass. In years with low macrophyte biomass (e.g., as observed
during recolonization in Lake V reng in 1990 [400 kg OW /ha in the autumn of 1989
and even lower in 1988]; see Fig. 20.1), an abundant waterfowl population is more
likely to have an impact. Furthermore, the impact of waterfowl may include more
than actual consumption, as both mute swan and coot tear up much larger quantities than they acutally consume (Berglund et al., 1963; Anderson and Low, 1976;
van Donk et al., 1994). Finally, indirect estimation of food consumption by using
bioenergetic models may yield significantly higher values, albeit that the uncertainty associated with such models is subject to debate (Mitchell and Wass, 1995).
Clausen and Krause-Jensen (unpubl.) thus estimated daily consumption for mute
swan and coot feeding on Zostera to be 530-680 g DW/day and 115-120 g
OW/day, respectively. Applying these figures to Lake Vreng yields a food consumption of 15-20 kg OW/day/ha at maximum waterfowl density, thus rendering
a possible impact of waterfowl more likely.
The colonization pattern recorded in Lake V reng lends further support to the
hypothesis that waterfowl inhibit colonization: first, colonization was delayed in
relation to the potential growth areas (here defined as areas with a depth less than
mean summer Secchi depth). Thus in 1988, less than 2% of the lake bottom was
covered with macrophytes, despite a potential growth area of about 50%. Second,
submerged macrophytes started to recolonize the wind-exposed and deepest parts
of the lake and not-as had been expected-the more shallow and sheltered areas
(Lauridsen et al., 1994). This colonization pattern probably reflects the impact of
grazing waterfowl because waterfowl mainly inhabit and nest in sheltered and
reed-covered areas during the summer and are much less abundant at windexposed and reed-poor locations (Lauridsen et aI., 1994).
Finally, caging studies confirmed the potential grazing effect of waterfowl. In
Lake Vreng, grazing by waterfowl led to decreased shoot length in unprotected
pots (Table 20.2), shoot length in protected pots at sheltered locations being 57.7
and 28.1 em on mud and sand, respectively, as compared with 14.8 and 9.9 em,
respectively, in unprotected pots. The 2.8-3.9-fold difference between protected
and unprotected pots at the sheltered location (compared with 1.8-2.1-fold difference at the exposed location) provides further evidence for a higher impact of
waterfowl grazing in sheltered areas.
Similar results were found in Lake Stigsholm. Total shoot length, mean shoot
length, and branches per shoot were significantly higher in protected pots than in
unprotected pots (Table 20.3). For instance, mean total shoot length was 15.7 m in
protected pots but only 3.7 m in unprotected pots. Moreover, the percentage of
304
M. S!lJndergaard et al.
Table 20.2. Mean Shoot Length (n = 8) of P. crispus Planted in BirdProtected and Unprotected Pots in Lake Vreng at Wind-Exposed and
Sheltered Locations on Mud and Sanda
Exposed, mud
Exposed, sand
Sheltered, mud
Sheltered, sand
Protected pots
cm (n)
Unprotected
pots cm (n)
P
67.6 (64)
48.4 (64)
57.7 (48)
28.1 (56)
38.6 (64)
22.6 (64)
14.8 (64)
9.9 (64)
.0007
.0004
<.0001
<.0001
aMean of six-eightdates from May 22 to August 10, 1989. Significant differences
between the two pot types during the period are indicated (GLM procedure using
protected or unprotected pots as class variables and shoot length and date as
dependent variables; SAS, 1990).
stubble, which is a direct index of grazing, was significantly higher in the unprotected pots (Table 20.3): 9.8% in protected pots versus 20.1 % in unprotected
pots. Furthennore, the percentage of stubble peaked at the beginning of July, when
coot density was highest (Sl1lndergaard et aI., 1996).
Although grazing by fish cannot be excluded in experiments of this kind (Prejs,
1984; van Donk et al., 1994), changing the experimental set-up in Lake Stigsholm
to allow fish (roach and perch) to enter the pots did not lead to any changes in
macrophyte growth. The impact by fish is therefore considered to be unimportant,
at least in this lake (S0ndergaard et aI., 1996).
It can thus be concluded that both submerged macrophyte coverage and waterfowl density are subject to marked interannual variation and that these variations
are closely and positively related. The experiments undertaken in Lake Vreng and
Lake Stigsholm suggest that waterfowl may suppress macrophyte biomass and
development in shallow lakes. The impact of waterfowl grazing is likely to be
particularly important (1) in lakes in which macrophytes are in the recolonization
phase after a reduction in nutrient loading, (2) in relatively small shallow lakes in
Table 20.3. P. crispus Shoot Characteristics in Bird-Protected and Unprotected Pots
(Mean of Eight Pots, Seven Locations, and Four Dates: July 27, August 10, August 24,
September 3, 1990) in Lake Stigsholma
Protected pots
Total shoot length (m)
Mean shoot length (cm)
Number of shoots
Branches per shoot
% Stubble
15.7
10.9
138
0.45
9.8
(SD = 10.6 )
(SD= 2.1 )
(SD = 82 )
(SD = 0.31)
(SD= 8.9 )
Unprotected pots
3.7
7.2
39
0.19
20.1
(SD=
(SD=
(SD =
(SD =
(SD =
0.57)
3.8 )
51 )
0.35)
18.4 )
P
<.001
<.001
<.001
.004
.010
aSignificant differences between the two pot types are indicated (GLM procedure as described in
Table 20.2).
20. Macrophyte-Waterfowl Interactions
305
which the impact of waterfowl is greater due to a relatively large littoral zone and
in which coot density in the absence of macrophytes is highest (Brl/lgger-Jensen
and JI/lrgensen, 1992), and (3) during the autumn/winter when they are not territorial. When attempting to achieve improved lake water quality, it may therefore
be a useful management strategy to protect sparsely developed macrophyte populations in the early phase of recolonization.
Acknowledgments. We wish to thank the technical staff of the National Environmental Research Institute for their assistance, in particular L. Hansen and J.
Stougaard-Pedersen, B. Laustsen, J. Glargaard, K. Jensen, and L. NI/lrgaard.
Layout and manuscript assistance was provided by K. MI/lgelvang, J. Jacobsen,
and A.M. Poulsen, and D. Barry provided useful editorial comments.
References
Anderson, M.G.; Low, J.B. Use of sago pondweed by waterfowl on the Delta Marsh,
Manitoba. J. Wild1. Manage. 20:233-242; 1976.
Berglund, B.E.; Curry-Lindahl, K.; Luther, H.; Olsson, V.; Rohde, W.; Sellerberg, G.
Ecological studies on the mute swan (Cygnus olor) in southern Sweden. Acta Vertebratica 2:167-288; 1963.
Brl/lgger-Jensen, S.; JI/lrgensen, H.E. Vandfugles og sl/lers miljl/ltilstand. Miljl/lprojekt 200 (in
Danish). [Waterfowl versus the environmental state of lakes. Environmental Project
200]. Copenhagen: Danish Environmental Protection Agency; 1992.
Clausen, P.; Krause-Jensen, D. An annual budget of eelgrass Zostera marina consumption
by herbivorous waterfowl in a shallow Danish estuary (in preparation).
Hargeby, A.; Andersson, G.; Blindow, I.; Johansson, S. Trophic web structure in a shallow
eutrophic lake during a dominance shift from phytoplankton to submerged macrophytes.
Hydrobiologia 2791280:83-90; 1994.
Hurter, H. Nahrungsokologie des Bllisshuhn (Fulica atra) an den Uberwinterungsgewiissem im
nordlichen Alpenvorland. Der Omitologische Beobachter 76:257-288; 1979.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; SI/lndergaard, M.; Mortensen, E.; Sortkjrer, 0.;
Olrik, K. Fish manipulation as a lake restoration tool in shallow, eutrophic, temperate
lakes. 2: Threshold levels, long-term stability and conclusions. Hydrobiologia 2001201:
219-227; 1990.
Jeppesen, E.; SI/lndergaard, M.; Kronvang, B.; Jensen, J.P.; Svendsen, L.M.; Lauridsen, T.
Lake and catchment management in Denmark. In: Harper, D.; Brierley, B.; Ferguson,
A.; Phillips, G.; Madgewick, J., eds. Ecological basis for lake and reservoir management. London: J. Wiley & Sons (in press).
Jupp, B.P.; Spence, D.H.N. Limitations of macrophytes in a eutrophic lake, Loch Leven, II:
Wave action, sediments and waterfowl grazing. J. Bco1. 65:431-446; 1977.
Kil/lrboe, T. Distribution and production of submerged macrophytes in Tipper Grund (Ringkl/lbing Fjord, Denmark), and the impact of waterfowl grazing. J. App1. Bco1. 17:675-687;
1980.
Lauridsen, T.L.; Jeppesen, E.; 0stergaard Andersen, F. Colonization of submerged macrophytes in shallow fish manipulated Lake Vreng: impact of sediment composition and
waterfowl grazing. Aquat. Bot. 46:1-15; 1993.
Lauridsen, T.L.; Jeppesen, E.; SI/lndergaard, M. Colonization and succession of submerged
macrophytes in shallow Lake Vreng during the first five years following fish manipUlation. Hydrobiologia 275/276:233-242; 1994.
Lodge, D. Herbivory on freshwater macrophytes. Aquat. Bot. 41: 195-224; 1991.
306
M.
S~ndergaard
et al.
Mitchell, S.; Wass, R.T. Food consumption and faecal deposition of plant nutrients by black
swan (Cygnus atratus Latham) in a shallow New Zealand lake. Hydrobiologia 306: 189197; 1995.
Prejs, A. Herbivory by temperate freshwater fishes and its consequences. Environ. BioI.
Fish. 10:281-296; 1984.
SAS. SAS language version 6. SAS Institute Inc., Cary, NC, USA; 1990.
Scheffer, M. Multiplicity of stable states in freshwater systems. Hydrobiologia, 200/201:
475-487; 1990.
Skotte-M~ller, H. Midqyllands Fugle. Omitologiske unders~gelser i Midtjylland (in Danish).
[The birds of central Jutland. Omitological investigations in central Jutland]. Meddelelse 2:34; 1970.
S~ndergaard, M.; Jeppesen, E.; Mortensen, E.; Dall, E.; Kristensen, P.; Sortkjrer, O. Phytoplankton biomass reduction after planktivorous fish reduction in a shallow, eutrophic
lake: a combined effect of reduced intemal P-loading and increased zooplankton grazing. Hydrobiologia 200/201:229-240; 1990.
S~ndergaard, M.; Bruun, L.; Lauridsen, T.L.; Jeppesen, E.; Vindbrek Madsen, T. The impact
of grazing waterfowl on submerged macrophytes: in situ experiments in a shallow
eutrophic lake. Aquat. Bot. 53:73-84; 1996.
van Donk, E.; De Deckere, E.; Klein Breteler, J.G.P.; Meulemans, 1. Herbivory by waterfowl and fish on macrophytes in a biomanipulated lake: effects on long-term recovery.
Verh. Int. Verein. Limnol. 25:2139-2143; 1994.
Winfield, I.J. Fishes, waterfowl and eutrophied ecosystems: a perspective from a European
vertebrate ecologist. In: Guissani, G.; Van Liere, L.; Moss, B., eds. Ecosystem research
in freshwater environment recovery. Mem. 1st. ItaI. IdrobioI. 48:113-126; 1991.
21. Influence of Macrophyte Structure, Nutritive
Value, and Chemistry on the Feeding Choices of a
Generalist Crayfish
Greg Cronin
Introduction
Herbivores are an important component of the food webs of nearly all communities that receive sunlight. The historical belief that freshwater macrophytes
enter the food web only as detritus is being challenged by recent evidence suggesting direct herbivory on live macrophytes can be significant (reviewed by Lodge,
1991; Newman, 1991; Lodge et aI., this volume, Chapter 8; Mitchell and Perrow,
this volume, Chapter 9). Most herbivory of freshwater macrophytes is caused by
generalist grazers such as aquatic insects (Newman, 1991; Jacobsen and Sand-Jensen, 1992, 1994, 1995), crustaceans (Chambers et aI., 1990; Lodge, 1991; Newman et aI., 1992; Creed, 1994; Lodge et aI., 1994), fish (Lodge et aI., this volume,
Chapter 8), waterfowl (Conover and Kania, 1994; Mitchell and Perrow, this
volume, Chapter 9), and mammals (Fraser et al., 1984; Doucet and Fryxell, 1993).
Little quantitative information exists about the feeding preferences of freshwater herbivores, and less is known about plant traits that detennine those preferences (Lodge, 1991; Newman, 1991), although macrophyte structure (Chambers
et aI., 1991; Jacobsen and Sand-Jensen, 1994), nutritive value (Lonnan, 1980;
Fraser et aI., 1984; Center and Wright, 1991; Doucet and Fryxell, 1993), and
secondary metabolites (Buchsbaum et aI., 1984; Newman et aI., 1992) have all
been implicated. As an example of our dismal mechanistic understanding of
freshwater macrophyte-herbivore interactions, thousands of secondary metabolites have been characterized from terrestrial and marine plants and the effects of
307
308
G. Cronin
dozens of these compounds on the feeding behavior of terrestrial (Rosenthal
and Berenbaum, 1992) and marine (Hay and Fenical, 1988; Paul, 1992) herbivores are known, but there is only one documented example of a specific
plant compound defending a freshwater macrophyte from herbivores (Newman et aI., 1992). Yet, there is no compelling reason to expect chemical
defenses to be rarer in freshwater than in terrestrial or marine plants (Ostrovsky and Zettler, 1986).
Crayfish are very common and important freshwater omnivores that include
much plant material in their diets. They can have large impacts on littoral and
stream communities by reducing the abundance of macrophytes and invertebrates
(Lodge and Lorman, 1987; Chambers et al., 1990; Creed, 1994; Lodge et at,
1994), yet little is known about their feeding selectivity or the effects of various
plant traits on their feeding decisions. In this study, I examined the feeding
preferences of the crayfish Procambarus clarkii among nine species of freshwater
macrophytes (including macroscopic algae) and measured their responses to
manipulation of the combined plant traits of morphology, toughness, and surface
features (hereafter grouped as "structure") and their response to plant extracts. I
also relate the preference of crayfish with measurements of plant nitrogen, protein,
and polyphenolic compounds.
Methods
The Organisms
Nine species of freshwater macrophytes that represented a wide range of growth
forms and taxonomic groups were collected in northern Indiana and southern
Michigan (all were collected within 50 km of the Notre Dame campus) on May 1,
1995. Five of the plants were submerged, including three algae (Chara sp. [Characeae,
Charophyta]; Spirogyra sp. [Zygnemataceae, Chlorophyta]; Batrachospermum sp.
[Batrachospermaceae, Rhodophyta]) and two angiosperms (Magnoliophyta) (Potamogeton amplifolius [Potamogetonaceae] and Ceratophyllum demersum [CeratophyllaceaeD. The remaining species were angiosperms with floating leaves (Nuphar
advena [Nymphaeaceae]) or emergent leaves (Typha lati/olia [Typhaceae], Iris virginica [Iridaceae], and Carex vesicariae [Cyperaceae D.
The Louisiana crayfish Procambarus clarkii is a commercially valuable crayfish that is cultivated far outside its native range of the southeastern United States
and now occurs as feral populations in Mexico, Japan, Hawaii, from coast to coast
in the continental United States (Hobbs, 1972), and as far north as Lake Erie.
Procambarus did not historically co-occur with all the macrophyte species used in
this study, but their current ranges overlap. Procambarus used in my experiments
were purchased from A1chafalaya Biological Supply Company (Louisiana) and
were kept individually in 25 x 31-cm plastic tubs with about 5 cm of untreated
well water. Crayfish used in these experiments had a carapace length of about 6 cm
and were maintained on commercial fish food between assays.
21. Feeding Choices of a Generalist Crayfish
309
Crayfish Preference Among Fresh Plants
Fresh pieces of the nine macrophytes were simultaneously offered to individual crayfish from May 1 to May 3, 1995, to determine their feeding
preferences. Within a plant species, each piece came from a different plant
shoot or clump of algae. I used pieces of plants that appeared to occupy similar
volumes of water in an attempt to provide similar encounter rates to the
crayfish. This procedure resulted in smaller masses of Potamogeton and Carex
(about 150 mg and 200 mg, respectively) than of the other species (about 600
mg each). The wet mass of each plant piece was determined by first spinning it
for about 7 seconds in a salad spinner to remove excess water and then
weighing it to the nearest milligram. A weighed piece of each of the nine
macrophyte species was placed in each of 20 tubs with crayfish. To control for
changes in mass not due to the crayfish, 11 tubs were set up in an identical
manner but without a crayfish. The pieces of plants were anchored to the
bottom of the plastic tubs with rubber suction cups.
After the crayfish had foraged for 1.5 days, the macrophytes were reweighed
and consumption was calculated from the equation [(Ho x CJCo) - Hf )]; where Ho
and Hf were the masses of the plants in each tub with a crayfish before and after
the assay, and Co and Cf were the mass of the macrophyte species from a randomly
chosen control tub before and after the assay. These latter measures accounted for
the 1-15% increase in mass unrelated to grazing activity with the following
average increases: Chara 1%; Ceratophyllum 3%; Spirogyra 4%; Typha 15%;
Potamogeton 1%; Batrachospermum 4%; Nuphar 9%; Carex 2%; and Iris 9%.
Floating and emergent leaves with large air spaces (Typha, Nuphar, Iris) showed
the largest increase in mass, presumably from absorbed water.
Crayfish Preference Among Reconstituted Plants
Additional plant tissue from each species was frozen, freeze-dried, ground into a
fine powder, and stored at -20 o e until used in other experiments. To determine if
plant structure influenced crayfish feeding decisions, the powdered macrophytes
were reconstituted into an alginate gel at natural dry mass concentration (i.e., the
amount of water added to the powdered macrophytes equaled the amount of water
removed by freeze-drying). To do this, a measured amount of powdered macrophyte (this amount varied among macrophyte species as percentage dry mass
varied among species; Table 21.1) was mixed into a 2% solution of sodium
alginate. This mixture was pressed into a mold to form a 2.6-cm-wide x 1.5-mmthick strip on a piece of fiberglass window screening material. A solution of 0.25
M calcium chloride was poured onto the macrophyte/sodium alginate mixture to
solidify it (sodium alginate is soluble in water; calcium alginate is a water-insoluble gel). The fiberglass screening material provided support for the reconstituted plants and a uniform grid that allowed me to quantify the amount eaten by
counting the squares of the screen that had been cleared of reconstituted plants
(see Fig. 1 in Hay et al. [1994] for an illustration of these methods). Thus, all
reconstituted macrophytes were of similar structure, but the nutritive value and
310
G. Cronin
Table 21.1. Tissue Properties of the Nine Macrophyte Species and Their Relation to Feeding
Preferencesa
% Dry
mass
Species
Chara
Ceratophyllum
Spirogyra
Typha
Potamogeton
Batrachospermum
Nuphar
Carex
Iris
Pearson correlation
coefficient
(whole-tissue assay)
Pearson correlation
coefficient
(reconstituted tissue)
Deterrency
of extract
(average
reduction in
feeding)
- 46
44
- 26
-109
Polyphenolics
(%WM)
Protein
(%WM)
Nitrogen
(%WM)
Carbon
(%WM)
0.22
0.34
0.11
0.42
0.35
0.19
0.62
0.78
0.45
-0.645
3.23
4.38
2.60
5.20
3.72
1.23
7.24
9.29
5.94
-0.530
0.11
-0.223
0.422
0.446
-0.213
10.7
11.4
6.5
12.4
9.1
3.0
15.8
21.3
13.4
-0.391
-10
93
45
44
-0.321
0.52
0.56
0.34
1.36
0.16
0.14
2.20
1.61
1.62
-0.642
0.574
-0.336
0.265
0.00
0.03
0.23
0.03
0.22
0.00
0.69
om
aThe concentrations of soluble protein, total nitrogen, and total carbon are given as the percentage of plant
wet mass. Order of species is same as in Figure 21.1.
taste of each reconstituted species were presumably little altered, although freezing and freeze-drying can affect some plant metabolites (Cronin et aI., 1995).
The strips of eight reconstituted plants (Potamogeton powder was unavailable
for this experiment) were cut into 1.2-cm sections, anchored to the bottom of tubs
with rubber suction cups, and simultaneously offered to 29 separate crayfish.
Because the reconstituted plants remained on the screening material in the absence
of crayfish, controls for autogenic changes were unnecessary.
Effects of Macrophyte Extracts on Crayfish Feeding
The combined effects of lipophilic and water-soluble metabolites on the feeding
behavior of crayfish were assessed by adding extracts to a standard palatable
artificial food. The standard food consisted of freeze-dried powdered Typha latifolia collected from Morehead City, North Carolina, distilled water, and alginate (1
g Typha powder: 7.3 rnl H 2 0: 167 mg sodium alginate).
Four grams of powdered macrophyte were extracted first with a 1: 1 mixture of
methanol and ethyl acetate to remove lipophilic metabolites and then with distilled
water to extract water-soluble metabolites. The lipophilic metabolites dissolved in
solvent were mixed with 4 g of powdered Typha, and the solvents were removed
by evaporation under reduced pressure. The sodium alginate was dissolved in the
aqueous extract, and this solution was mixed with the Typha powder containing
21. Feeding Choices of a Generalist Crayfish
311
the lipophilic extract. This mixture was spread onto fiberglass screening and
solidified into a gel as described above. Standard food treated in the same manner,
but without added extracts, was used as control food. A parallel strip of control
food was made 1 cm from the treated food on the same piece of fiberglass screen.
The screen with paired treated and control food was cut into "test strips," each
consisting of two rectangles of artificial food that differed only in the presence of
a macrophyte crude extract in one of the pieces of food. These experiments were
performed on separate days for each macrophyte species except Potamogeton.
Test strips were offered to individual crayfish for 1-12 hours (depending on
their feeding rates), and the amount of food consumed was quantified by counting
the number of squares in the screen that had been cleared of food. Replicates in
which none or all of both food types were eaten were not used in data analysis.
Data for each feeding assay were analyzed with a paired-sample t-test.
Plant Tissue Constituents
Powdered macrophytes were analyzed for total nitrogen and carbon with a Perkin
Elmer eN analyzer (model 2400), soluble protein with a modified Bradford assay
(Duffy and Hay, 1991), and polyphenolics with a modified Folin-Denis assay
(Yates and Peckol, 1993).
For polyphenolic analyses, about 5 mg of ground, freeze-dried tissue, weighed
to the nearest 0.1 mg, was extracted with 1.00 ml of 50% aqueous methanol for 1
day at 1"C with occasional mixing. Two 100-!!1 aliquots of the extraction solution
were placed in separate test tubes and diluted with 8.4 ml of acidic 10% methanol
(pH = 2). Polyvinylpolypyrrolidone (PVPP; Sigma) was added to one test tube.
The Folin-Denis reagent reacts with several chemical constituents, including
phenolics, proteins, amino acids, and ascorbic acid (Andersen and Todd, 1968),
whereas PVPP binds specifically with polyphenolics, preventing them from reacting with the Folin-Denis reagent. The test tube without PVPP provided the traditional Folin-Denis reading that included phenolics, protein, amino acids, etc., and
the +PVPP test tube provided a Folin-Denis reading for everything except polyphenolics. The concentration of polyphenolics was calculated from the difference
in the two readings, using tannic acid as a standard.
Results
When plants were offered as fresh tissue, Procambarus consumed large amounts of
Chara, Ceratophyllum, and Spirogyra; intermediate amounts of Typha; and little (if
any) of Potamogeton, Batrachospermum, Nuphar, Carex, or Iris (Fig. 21.1A). Morphologically, the three preferred species are finely branched or filamentous. Except for
Batrachospermum, the less preferred species have flat (Potamogeton and Nuphar) or
blade-like (Typha, Carex, and Iris) leaves. Batrachospermum is very flaccid and
covered in copious slime.
When plants were offered as reconstituted tissue in an alginate gel, the feeding
preferences of the crayfish were altered (compare Fig. 2 1.1 A and B). Large
312
G. Cronin
A 600
Figure 21.1. Amount of plants
consumed by Procambarus clarkii
among macrophyte species when
offered simultaneously (A) as fresh
tissue and (B) as freeze-dried, powdered macrophytes that had been
reconstituted with an alginate gel.
(C) The effects of crude extracts
from the macrophyte species on
feeding by the crayfish. Bars represent mean (+ 1 SE). Each pair of
bars in C represents an independent
feeding assay, with the sample size
given below the pair. An asterisk
indicates that the means in a pair
differ significantly from each other
at a =0.05.
Fresh Plants
N=20
Cl
S400
c:
*Q)
§
o
200
E
«
o
B
120
Reconstituted Plants
N=29
c:
~
80
~
ro
5-
(/)
40
o
Plant Extracts
*
c:
$
ro
60
D-controi
~ - treated with extract
*
I
UJ
*
01<
Ul
~
ro
::J
& 30
'o"'J
"Cl
0
0::
~
I
amounts of Chara, Typha, and Carex; intermediate amounts of Ceratophyllum and
Nuphar; and little Spirogyra, Batrachospermum, and Iris were consumed (Fig.
21.1B). Thus, plant structural properties are important determinants of Procambarus feeding decisions because altering the structure changed preferences. However, additional plant traits are also important given the crayfish fed selectively
among the structurally identical reconstituted plants.
21. Feeding Choices of a Generalist Crayfish
313
Crude extracts from plants contain both feeding stimulants and deterrents.
When added to the standard food, the crude extracts from Ceratophyllum, Nuphar,
and Iris significantly reduced consumption by 44%, 93%, and 44%, respectively.
By contrast, the extract from Typha significantly stimulated feeding by 109%
(Table 21.1; Fig. 21.1C). Extracts of the remaining plants did not significantly
affect crayfish feeding. Thus, plant chemistry can deter or stimulate feeding by
crayfish. The deterrency of the crude extracts, as determined by the average
percentage reduction in feeding caused by the extract (Fig. 21.IC), was negatively
related to the amount of whole tissue and reconstituted tissue consumed during
feeding assays (Table 21.1).
No measured plant trait was significantly correlated with feeding preferences,
either as whole plant tissue or reconstituted tissue. However, I provide correlation
coefficients in Table 21.1 to indicate feeding trends with respect to tissue traits.
Polyphenolics, a class of secondary metabolites with putative defensive functions, were detected at various levels in all the angiosperms, were undetectable in
Chara and Batrachospermum, and surprisingly composed 0.23% of the wet mass
(=3.5% dry mass) of the green alga Spirogyra (Table 21.1). Nuclear Magnetic
Resonance analysis of a methanol extract of Spirogyra revealed the presence of
aromatic phenolic compounds, but these compounds differed from brown algal
phlorotannins or plant tannins (W. Fenical, personal communication). Polyphenolics were negatively related to the amount of tissue consumed in both the
whole-tissue assay and the reconstituted-tissue assay (Table 21.1).
Nitrogen and protein varied considerably among species and were not well
correlated when expressed as percentage of dry mass (Pearson correlation coefficient = 0.044; P = .991; n =9). However, because of the major influence of water,
which constituted 78-97% ofthe plants wet mass, protein and nitrogen concentrations were significantly correlated with each other when expressed on a wet mass
basis (Pearson correlation coefficient = 0.818; P = .007; n =9). On a percentage
dry mass basis, the concentrations of nitrogen and protein tended to be lower in the
submerged plants (Chara, Ceratophyllum, Spirogyra, Potamogeton, and Batrachospermum; average 3.3% nitrogen and 4.3% protein) than in the floating (Nuphar; 3.9% nitrogen and 14.0% protein), emergent (Typha; 3.4% nitrogen and
11.0% protein), or shore (Carex and Iris; average 3.6% nitrogen and 9.9% protein)
species. This pattern is exaggerated if concentrations are expressed on a percentage wet mass basis because submerged tissues have a higher water content than
emergent tissues (Table 21.1). Nitrogen and protein concentrations were negatively related with the feeding preferences of crayfish during the whole-tissue assay
but were positively related to the feeding preference of crayfish during the reconstituted-plant assay (Table 21.1).
Discussion
Generalist herbivores, including Procambarus, base their feeding decisions on
multiple plant traits such as morphology, structure, chemical defenses, and nutri-
314
G. Cronin
tive value (Lodge, 1991; Newman, 1991; Lodge et aI., this volume, Chapter 8).
Because herbivores are generally more limited by nitrogen than carbon (Mattson,
1980), nitrogen or protein have been considered important feeding stimulants for
these animals. Apparently, Procambarus fed on plants such as Chara, Ceratophyltum, and Spirogyra when offered as whole tissue because their finely branched or
filamentous morphologies make them easier to handle, shred, and consume. When
the difficulties imposed by plant structure were removed by forming all plants into
a gel, the feeding preferences of Procambarus were altered; some plants with high
concentrations of protein, nitrogen, and dry mass (i.e., nutritious) such as Typha
and Carex became more popular food items while some less nutritious plants such
as Ceratophyllum and Spirogyra became less popular (Fig. 21.1B).
Nutritious macrophytes that were low preference as whole and reconstituted
plants (Nuphar and Iris) contained chemical defenses (Fig. 21.1 C). The preferences for whole and reconstituted plants were both negatively related to the deterrency of the crude extract and concentration of polyphenolics (Table 21.1). Although
some plant chemical defenses reduced crayfish feeding significantly, they were
not entirely effective given that a moderate amount of reconstituted Nuphar was
consumed (Fig. 21.1B). Additionally, Ceratophyllum had a deterrent crude extract,
yet it was highly preferred as fresh whole tissue and moderately preferred as
reconstituted tissue. Other plant traits, such as a slimy surface, may also deter
crayfish: Batrachospermum was low preference despite a very flaccid morphology and lack of a deterrent extract, but it was extremely slimy. However, it also
had little nutritive value (i.e., it was 97% water).
Procambarus will eat only plants that they can handle, shred, and ingest.
Although this is self-evident, it helps explain why feeding preferences during the
whole-tissue assay were negatively related to plant traits normally considered to
be feeding stimulants such as protein, nitrogen, and dry mass. Plants with high
concentrations of dry mass, and hence high concentrations of nutrients on a wet
mass basis, also had high amounts of structural material. Only after the plants were
made structurally identical were nutritive plant traits positively related to feeding
preferences (Table 21.1). The importance of plant structure in determining its
susceptibility to grazing has been previously noted for seaweeds; hard encrusting
forms are among the least susceptible to herbivores whereas highly branched or
filamentous forms are generally the most susceptible (Littler and Littler 1980).
However, even seaweeds with susceptible forms can reduce grazing damage by
being chemically defended (Cronin and Hay, 1996a,b). Among plants with similar
levels of defense, palatability should be positively related to nutritive value
(Lodge et al., this volume, Chapter 8). For the five reconstituted plants that were
not chemically defended, the crayfish ate nearly all the food made from the three
species (Chara, Typha, and Carex) with the highest concentration of protein and
total nitrogen (wet mass basis), whereas food made from the less nutritious species
(Spirogyra and Batrachospermum) was consumed less. Chambers et al. (1991)
found that the feeding preferences of the crayfish Orconectes virilis were negatively correlated with plant nutritive value, probably because the less nutritious
plants were easier to handle. Reduction in the structural integrity of macrophytes
21. Feeding Choices of a Generalist Crayfish
315
after death helps explain why macrophytes are consumed more as detritus than
living tissue, an explanation that receives less attention than the "microbial conditioning" or "chemical defense leaching" explanations (Newman, 1991).
Other factors not considered above, such as cover or protection from predators
afforded by the plant (Damman, 1987; Duffy and Hay, 1994), the consumer's state
of hunger (Cronin and Hay, 1996b), or the consumer's prior feeding experiences
(Provenza, 1995; Howard et al., 1996), will also affect herbivore feeding decisions. Although a single plant trait (e.g., thick, tough, blade-like leaf structure)
can be used to accurately predict a plant's susceptibility to Procambarus, it may
not be useful to predict a plant's susceptibility to herbivory in general because the
effects of plant traits on feeding vary among herbivore species. Additionally,
multiple plant traits can interact to influence herbivore feeding behavior in a
mitigative, additive, or synergistic manner (Duffy and Paul, 1992; Hay et al.,
1994).
It has been demonstrated that the feeding activity of generalists can have
significant direct and indirect impacts on aquatic communities (Lodge and Lorman, 1987; Chambers et al., 1990; Creed, 1994; Lodge et aI., 1994; Hill and
Lodge, 1995; Lodge et al., this volume, Chapter 8; Mitchell and Perrow, this
volume, Chapter 9). A mechanistic understanding of macrophyte-herbivore interactions will improve our ability to predict the impacts of herbivores on freshwater
communities. We know that herbivores can reduce the standing stock of macrophytes, but a knowledge of herbivore feeding preferences lmd plant defensive
traits will allow better herbivore-specific and macrophyte-specific predictions of
impact.
Summary
Laboratory experiments were perfonned to determine the feeding preferences of
the crayfish Procambarus clarkii among nine species of submerged, floating,
emergent, and shoreline macrophytes representing a broad taxonomic range (9
families in 4 divisions). When plants were offered as whole fresh tissue, the
crayfish preferred highly-branched and filamentous plants over those with thick,
flat leaves. However, when the plants were freeze-dried, ground into a powder,
and reconstituted with an alginate gel at the original percentage of dry mass (i.e.,
plant structure was equalized, but nutritive value and taste were presumably
unaltered), the feeding preferences of the crayfish were greatly altered. Chemical
extracts from the selected plant species that were incorporated into palatable
artificial foods also altered consumption by the crayfish: crude extracts of three
macrophytes deterred feeding while the extract of one macrophyte significantly
stimulated feeding. Therefore, plant structure (morphology, toughness, and/or
surface features) and plant chemistry are important deterrnimmts of crayfish feeding choices. Finally, total nitrogen, protein, and polyphenolics were quantified
from plant tissues to relate to feeding preferences. No single plant trait could
explain the feeding preferences of crayfish. Rather, crayfish apparently base
316
G. Cronin
feeding decisions on a variety of plant traits: Procambarus avoided species
with structural or chemical deterrents, and preferred undefended plants high in
nitrogen.
Acknowledgments. This research was funded by NSF DEB 94-08452-002 to
David Lodge. I am grateful to Bill Fenical for performing NMR studies, Jill
Witkowski for technical assistance, and David Lodge for insightful discussions
and comments about this project and herbivory in general. Reviews by Robin
Bolser, David Lodge, and two anonymous reviewers improved the manuscript.
References
Andersen, RA; Todd, J.R. Estimation of total tobacco phenols by their bonding to polyvinylpyrrolidone. Tobacco Sci. 12: 107-111; 1968.
Buchsbaum, R.; Valie1a, I.; Swain, T. The role of phenolic compounds and other plant
constituents in feeding by Canada geese in a coastal marsh. Oecologia 63:343-349;
1984.
Center, T.D.; Wright, AD. Age and phytochemical composition of waterhyacinth (Pontederiaceae) leaves determine their acceptability to Neochetina eichhorniae (Coleoptera:
Curculionidae). Environ. EntomoI. 20:323-334; 1991.
Chambers, P.A; Hanson, J.M.; Burke, J.M.; Prepas, E.E. The impact of the crayfish
Orconectes virilis on aquatic macrophytes. Freshwat. BioI. 24:81-91; 1990.
Chambers, P.A.; Hanson, J.M.; Prepas, E.E. The effect of aquatic plant chemistry and
morphology on feeding selectivity by the crayfish, Orconectes virilis. Freshwat. BioI.
25:339-348; 1991.
Conover, M.R; Kania, G.S. Impact of interspecific aggression and herbivory by mute
swans on native waterfowl and aquatic vegetation in New England. Auk 111:744-748;
1994.
Creed, R.P., Jr. Direct and indirect effects of crayfish grazing in a stream community.
Ecology 75:2091-2103; 1994.
Cronin, G.; Hay, M.E. Amphipod grazing induces chemical defenses in the brown alga
Dictyota menstrualis. Ecology 77:2287-2301; 1996a.
Cronin, G.; Hay, M.E. Seaweed-herbivore interactions depend on recent history of both the
plant and animal. Ecology 77: 1531-1543; 1996b.
Cronin, G.; Lindquist, N.; Hay, M.E.; Fenical, W. Effects of storage and extraction procedures on yields of lipophilic metabolites from the brown seaweeds Dictyota ciliolata and
D. menstrua lis. Mar. EcoI. Prog. Ser. 119:265-273; 1995.
Damman, H. Leaf quality and enemy avoidance by the larvae of a pyralid moth. Ecology
68:88-97; 1987.
Doucet, C.M.; Fryxell, 1M. The effect of nutritional quality on forage preference by
beavers. Oikos 67:201-208; 1993.
Duffy, J.E.; Hay, M.E. Food and shelter as determinants of food choice by an herbivorous
marine amphipod. Ecology 72:1286-1298; 1991.
Duffy, J.E.; Hay, M.E. Herbivore resistance to seaweed chemical defense: the roles of
mobility and predator risk. Ecology 75:1304-1319; 1994.
Duffy, J.E.; Paul, v.J. Prey nutritional quality and the effectiveness of chemical defenses
against tropical reef fishes. Oecologia 90:333-339; 1992.
Fraser, D.; Chavez, E.R; Paloheimo, J.E. Aquatic feeding by moose: selection of plant
species and feeding areas in relation to chemical composition and characteristics of
lakes. Can. J. ZooI. 62:80-87; 1984.
Hay, M.E.; Fenical, W. Marine plant-herbivore interactions: the ecology of chemical
defense. Annu. Rev. EcoI. Syst. 19:111-145; 1988.
21. Feeding Choices of a Generalist Crayfish
317
Hay, M.E.; Kappel, Q.E.; Fenical, W. Synergisms in plant defenses against herbivores:
interactions of chemistry, calcification, and plant quality. Ecology 75: 1714-1726; 1994.
Hill, A.M.; Lodge, D.M. Multi-trophic-level impact of sublethal interactions between bass
and omnivorous crayfish. J. North Am. Benth. Soc. 14:306--314; 1995.
Hobbs, H.H., Jr. Crayfishes (Astacidae) of North and Middle America. U.S. Environmental
Protection Agency, Washington; 1972.
Howard, J.J.; Henneman, M.L.; Cronin, G.; Fox, J.A.; Hormiga, G. Conditioning of scouts
and recruits during foraging by a leaf-cutting ant, Atta colombica. Anim. Behav. 52:
299-306; 1996.
Jacobsen, D.; Sand-Jensen, K. Herbivory of invertebrates on submerged macrophytes from
Danish freshwaters. Freshwat. BioI. 28:301-308; 1992.
Jacobsen, D.; Sand-Jensen, K. Invertebrate herbivory on the submerged macrophyte Potamogeton perfoliatus in a Danish stream. Freshwat. BioI. 31 :43-·52; 1994.
Jacobsen, D.; Sand-Jensen, K. Variability of invertebrate herbivory on the submerged
macrophyte Potamogeton perfoliatus. Freshwat. BioI. 34:357-365; 1995.
Littler, M.M.; Littler, D.S. The evolution of thallus form and survival strategies in benthic
marine macroalgae: field and laboratory tests of a functional form model. Am. Nat.
116:25-44; 1980.
Lodge, D.M. Herbivory on freshwater macrophytes. Aquat. Bot. 41:195-224; 1991.
Lodge, D.M.; Lorman, J.G. Reductions in submersed macrophyte biomass and species
richness by the crayfish Orconectes rusticus. Can. 1. Fish. Aquat. Sci. 44:591-597;
1987.
Lodge, D.M.; Kershner, M.W.; Aloi, J.E.; Covich, A.P. Direct and indirect effects of an
omnivorous crayfish (Orconectes rusticus) on a freshwater littoral food web. Ecology
75:1265-1281; 1994.
Lorman, J.G. Ecology of the crayfish Orconectes rusticus in northern Wisconsin. Ph.D.
thesis, University of Wisconsin, Madison; 1980.
Mattson, W.J. Herbivory in relation to plant nitrogen. Annu. Rev. Ecol. Syst. 11:119-161;
1980.
Newman, R.M. Herbivory and detritivory on freshwater macrophytes by invertebrates: a
review. J. North Am. Benth. Soc. 10:89-114; 1991.
Newman, R.M.; Hanscom, Z.; Kerfoot, W.e. The watercress glucosinolate-myrosinase
system: a feeding deterrent to caddisflies, snails and amphipods. Oecologia 92: 1-7;
1992.
Ostrovsky, M.L.; Zettler, E.R. Chemical defenses in aquatic plants. 1. Ecol. 74:279-287;
1986.
Paul, V.J., ed. Ecological roles of marine natural products. Ithaca, NY: Comstock Publishing Associates; 1992.
Provenza, ED. Tracking variable environments: there is more than one kind of memory. J.
Chern. Ecol. 21:911-924; 1995.
Rosenthal, G.A.; Berenbaum, M.R., eds. Herbivores: their interactions with secondary plant
metabolites. Vol. II: Evolutionary and ecological processes. New York: Academic Press;
1992.
Yates, J.L.; Peckol, P. Effects of nutrient availability and herbivory on polyphenolics in the
seaweed Fucus vesiculosus. Ecology 74:1757-1766; 1993.
22. Concordance of Phosphorus Limitation in Lakes:
Bacterioplankton, Phytoplankton, Epiphyte-Snail
Consumers, and Rooted Macrophytes
Robert E. Moeller, Robert G. Wetzel, and Craig W. Osenberg
The characterization of many unpolluted lakes as aquatic "deserts" (Whittaker,
1975) reflects the well-recognized role of phosphorus and nitrogen in controlling
the abundance and growth rates of planktonic algae (Schindler, 1974, 1978; Kalff
and Knoechel, 1978). Deficiencies or reduced availability of these elements may
be less severe in the littoral zone because of marked morphological and physiological adaptations that result in metabolic and community mechanisms for nutrient recycling and retention (Wetzel, 1990a, 1993). As a result, few limiting
nutrient studies have been performed in the littoral zone, and many are directed
toward specific components (e.g., Fairchild and Everett, 1988; Fairchild and
Sherman, 1993).
In Lawrence Lake, located in southwest Michigan, submerged macrophytes are
responsible for less than one-quarter of the annual net primary productivity of
180 g C/m2 • The rest originates from phytoplanktonic (16%) and epiphytic (70%)
algae (Burkholder and Wetzel, 1989). To evaluate the extent of phosphorus limitation in these three communities as well as the bacterioplankton, we provided local
additions of phosphate separately to subsamples of each community. For comparability within the natural light and temperature regimes, all results are for
vegetation and microbial communities at a depth of 2 m. Rooted vascular plants
grow as deep as 6 m in this 12-m deep lake (Rich et aI., 1971; Wetzel et al., 1972;
Burkholder and Wetzel, 1989).
Phytoplankton assays consistently demonstrated a deficiency of phosphorus
during most of the year (Wetzel, 1981). Addition of 100 /lg P/L to samples of
318
22. Phosphorus Limitation in Lakes
en
:s
--
:J
OJ
'":1
--
1000
14C UPTAKE
100
319
.r···· .1
.' .
10y-----~
100 __- - - - - - - - d
PHYTOPLANKTON
10
BIOMASS ....
f .......•
as
...J
J:
o
o
3
6
9
DAYS
Figure 22.1. Progressive response of summer phytoplankton to P additions. Relative rates
of photosynthesis (upper, as relative 14C incorporation, in photosynthesis, as disintegrations
per second, DPS) and biomass (lower, as /lg chlorphyll aIL) for enriched (e, .A) and control
cultures (0, ~). Values are means ± 95% CI for n = 7 dates from June 7 to September 13.
Phosphate addition was 50 Ilg P/500 ml Pyrex flask, as K2 HP04. Cultures were maintained
on a rotating tray at ambient lake temperatures and light (16 h light/8 h dark) (Wetzel, 1981).
Chlorophyll a was corrected for phaeopigments (Wetzel and Likens, 1991). The 14C activity
of filtered algae from short-term labeling assays is a relative measure of photosynthesis:
incubation times, volumes filtered, and specific activities were consistent throughout.
Similar results were obtained with enrichments by organic phosphorus as ~-glycerophos­
phate (Wetzel, 1981).
lakewater incubated in the laboratory caused consistent increases in phytoplankton biomass and photosynthesis over the next 9 days (Fig. 22.1). Alkaline
phosphatase activity of enriched cultures increased at 9 days, after about 50%
suppression of activity at 3-6 days, which suggested a retum to phosphoruslimited conditions (Wetzel, 1981). The lO-fold stimulation of biomass of phytoplankton demonstrated a strongly limiting role for phosphorus under ambient
planktonic conditions, which included abundant dissolved nitrate (nitrate averaged
30/-lM and total dissolved P only 0.1-0.3 11M in summer at 2 m). Phytoplankton
development in 1984, when the littoral vegetation was studied, was very similar to
that in 1976, as had been the condition when the phytoplankton enrichments were
conducted. Variations in the annual phytoplanktonic mean chlorophyll a concentrations and annual in situ rates of photosynthesis varied less than 15% over an
18-year period of continuous measurement (Wetzel, 1983).
In situ rates of productivity of bacterioplankton from !the same sites were
determined by incorporation rates of tritiated thymidine into DNA (Wetzel and
Likens, 1991). Growth rates were also evaluated by changes in frequency of cell
division. Nutrient enrichment experiments with bacterioplankton also indicated an
enhanced rate of growth in response to phosphorus (Coveney and Wetzel, 1988,
1992, 1995).
320
R.E. Moeller, R.G. Wetzel, and C.w. Osenberg
A continuous and localized addition of phosphorus to the natural epiphyte
community was accomplished during July-September 1984 in a 2-m-diameter site
(plot A) dominated by a submerged sedge, Scirpus subterminalis Torr., the dominant macrophyte of the lake (Rich et aI., 1971; Wetzel et aI., 1972). Vertical rods
(41 rods, 1.7 g P/rod positioned on July 13; 5 rods added near center on August 20)
coated with resin-encapsulated pellets of calcium phosphate (Sierra Chemical Co.,
Milpitas, CA) were positioned 60 cm above the sediments throughout the plot;
phosphate released from the pellets over 3 months was at a rate of 0.5-1 % of the
annual loading of phosphorus to the lake. At a nearby site (plot B), rods coated
with pellets of mixed fertilizer (supplying N, P, K, S, and Ca) were embedded in
the sediment (30 cm below the surface (104 rods, 0.5 g P/rod on June 12) with 60%
of the rods 0.5 m from center of plots). Plot B was an attempt to stimulate
phosphorus release from growing macrophyte tissue, which is known to be a direct
although variable source of phosphorus for overlying epiphytes (Carignan and
Kalff, 1982; Moeller et al., 1988; Burkholder and Wetzel, 1990).
A visible increase in epiphytes occurred within 5 days when phosphate was
released above the sediment. After 10 weeks, we measured a 40-fold increase of
epiphyte biomass at the center of plot A compared with biomass immediately
outside the plot (Fig. 22.2). Epiphyte biomass was unchanged by the belowsedimentenrichment at plot B, although additional phosphorus was incorporated
by Scirpus. Leaf phosphorus increased from 0.14 to 0.51 % of dry wt within the
plot (P <.001). Evidently, the conservative retention of phosphorus by growing
macrophytes (Carignan and Kalff, 1982; Moeller et aI., 1988; Burkholder and
Wetzel, 1990; Wetzel, 1990a) was not significantly relaxed as excess phosphorus
accumulated. In plot A, the local addition of phosphorus to the strongly phosphorus-deficient natural epiphyte vegetation led to a massive algal proliferation
supported by nitrate, dissolved silica, and other nutrients continuously transported
into the unenclosed site.
Epiphyte biomass increased in plot A despite an increase in the intensity of
grazing by gastropods (Fig. 22.2). Snail biomass per unit area of lake bottom
increased 73% (115 vs. 66 mg dry wt/0.05 m2 ; P <.05). Snail biomass per unit of
macrophyte biomass increased 46% (34 vs. 23 mg dry wt/g organic wt; P <.1). The
response was attributable to two prosobranchs, Amnicola limnosa and Valvata
tricarinata, which feed on epiphytic microflora and detritus and made up more
than 80% of the total snail biomass. Both species live only 1 year and had
reproduced before the enrichment started. Snail densities per unit area of lake
bottom or unit of macrophyte biomass did not change (P >.2), indicating that
neither immigration nor reproduction significantly contributed to the response.
Instead, individuals of both common species were significantly larger (P <.005)
within the zone of increased epiphyte biomass (Fig. 22.2), which demonstrated
that snail growth rates responded positively to an increased food supply. Subsequent experiments replicated these results for epiphytes and snails in 2 additional years (Osenberg, 1988, 1989). Under natural conditions, therefore, food-limited
grazers fed on phosphorus-limited algae. The intensity of grazing may have been
sufficient to reduce algal biomass secondarily, as demonstrated elsewhere for
22. Phosphorus Limitation in Lakes
0.3
>z
0
VALVATA
AMNICOLA
102
lao
0.2
321
w
::>
0
w 0.1
a:
u.
0
0
1
U)
::>
0..
a:
10
(3
'~'"
U)
.....01
!as
0
1
2
SNAIL WEIGHT (mg)
,...
0.1
I
EPIPHYTE
BIOMASS
l!!
k' I
j
2
3
..J
:I:
0
0
DISTANCE (m)
Figure 22.2. Response of epiphytic algae and snails to 10 weeks of P addition. Mean epiphyte
biomass (± 95% CI, n == 7) increased within water-column enriched sites (....) but not at
sediment-enriched sites (~). Weight-frequency distributions ofthe two most common snails were
shifted to larger sizes in the enriched zone (e == enriched zone <m from center of plot, 0 == control
zone 2 m from center; n == 56-203 snails). Epiphytes were delicately brushed and rubbed from
entire tillers of Scirpus collected using SCUBA, then centrifuged, lyophilized, and extracted by
grinding in 90% basic acetone. Chlorophyll a was corrected for phaeopigment degradation
products (Wetzel and Likens, 1991). For epiphytes, seven replicates were collected between
September 18 and October 3 from each of nine distances from centers of plots. Snails were
collected in mid-October from 0.05-m2 quadrats of surficial sediment with overlying macrophytes; n == 4 enriched quadrats <m from center of plot A and 4 control quadrats >2 m from
center. New growth of Scirpus was reduced by heavy epiphyte load inside the enrichment
zone, so five-six entire stems of Potamogeton illinoiensis were also collected both inside
and outside the enrichment zone to better compare snail biomass per unit macrophyte. Shell
lengths were measured and converted to tissue dry mass based on regressions of lengthmass.
littoral (Cattaneo, 1983; Cuker, 1983; Lowe and Hunter, 1988; Osenberg, 1988,
1989; Bronmark, 1989; Lodge et al., 1994) or pelagic algal communities (Carpenter and Kitchell, 1984; Wright and Shapiro, 1984; Lehman and Sandgren,
1985). Our nutrient enrichments were intentionally of sufficiently short duration
that phytoplankton and epiphytes could outgrow their grazer populations, which
revealed the primary underlying control by phosphorus.
Aquatic macrophytes respond more slowly than rnicroalgae to changing nutrient availability, so we allowed a full year between initial enrichment and final
determination of biomass at five sites around the lake. Each site included 10
R.E. Moeller, R.G. Wetzel, and c.w. Osenberg
322
(/)
::>
0
0:
::r
Q.
0.4
0.3
.-_.. -f-·f· _. -------f ---
(/)-.
O~ 0.2
::r
.....
Q.
u.
0.1
...J
0
200
<
w
(/)
(/)
< ....
~
_ E
0"
Ill .....
01 100
u. .....
<
w
...J
0
SITES(N):
0
A-C(30)
N NP
Figure 22.3. Response of the dominant
submerged macrophyte Scirpus subterminaUs to 1 year of P addition. Upper: P
concentration in leaf tissue (% carbonatefree dry wt). Lower: Leaf biomass (g dry
wtlm2). Means with 95% CI are cornpared for three sites with a parallel Palone enrichment and for two sites with
a N-alone enrichment (0 =control, NP =
mixed fertilizer, N = +nitrogen, P =
+phosphorus). Fertilizer rods were as in
Figure 22.2, one rod/quadrat on each
date; nitrogen rods carried pellets of ammonium nitrate with some ammonium
sulfate (2.6 g N/rod). F-tests for site
effects were insignificant (P >.2) for
all treatments; thus, quadrats were treated
as replicates.
D-E(20)
control quadrats 0.008 m 2 in area, 10 quadrats receiving mixed fertilizer in
November 1983 and June 1984, and 10 separate quadrats receiving either calcium
phosphate or ammonium nitrate on the same dates (Fig. 22.3). Because rooted
macrophytes obtain most of their phosphorus from the sediment in nutrient-poor
waterbodies (Barko and Smart, 1980; Carignan & Kalff, 1980; Carignan, 1982;
Wetzel, 1990b), we avoided a proliferation of epiphytes by inserting fertilizer
beneath the sediment surface.
The biomass of the wintergreen leaves of Scirpus subterminalis as measured in
late October 1984 represents much of their annual net production (Rich et aI.,
1971; Wetzel et aI., 1972; Moeller and Wetzel, unpublished data). This biomass
was homogeneous among sites within treatments. The aggregated data (Fig. 22.3)
displayed a statistically significant increase caused by mixed fertilizer (P <.01)
and by phosphorus alone (P = .05). Nitrogen added without phosphorus had no
effect (P >.2). Potamogeton illinoiensis Morong made up 30% of total macrophyte
biomass at control sites and apparently responded similarly to the mixed fertilizer.
The increase from 44 to 60 g dry wtlm 2 for this more heterogeneously distributed
plant was not statistically significant (P >.1).
The phosphorus additions effectively saturated plant demand for that element,
as demonstrated by the accumulation of excess phosphorus in leaves (Fig. 22.3).
The small size of the subsequent increase in biomass (about 50%) probably means
that only the lowest leaf phosphorus concentrations encountered in Lawrence
Lake (e.g., the 27% of analyses <0.12% of dry wt) truly represent phosphorus
deficiency. Concentrations of nitrogen and potassium did not increase after addition of mixed or nitrogen-only fertilizers, which suggested that the enrichments
may not have increased their availability. Therefore, it is possible that these
22. Phosphorus Limitation in Lakes
323
elements may also have a limiting role unresolved by the less effective enrichments in these elements.
In the emergent macrophyte wetland surrounding this lake, a dominant species
is the cornrnon cattail Typha tatifolia L. Detailed analyses of tissue nutrient
analyses and population and growth dynamics in response to fertilization experiments revealed that T. tati/olia was principally phosphorus-limited in the open
calcareous marshes along the eastern side of the lake (Grace and Wetzel, 1981;
Dickerman and Wetzel, 1985). Populations of T. tatifolia under shaded conditions
within wooded areas of the western side of the lake were light rather than nutrientlimited (Grace and Wetzel, 1981).
Qualitatively, then, phosphorus operates concordantly as a growth-regulating
factor across all biotic components examined. Quantitatively, however, phosphorus is moderately available in the anoxic calcareous sediments of Lawrence
Lake that allows modest development of macrovegetation. Experiments with
potted Scirpus grown in the lake showed that biomass can be more than doubled to
500 g dry wtlm2 within one growing season; physical disturbance plus mixed
fertilizer, not phosphorus alone, was required. The heavily developed epiphytic
algal component of the littoral zone shares a phosphorus-deficient medium with
the phytoplankton. Rooted macrophytes occupying the same spatial habitat as
their epiphytes are functionally detached, by the nutrient geochemistry of sediments, from the aquatic desert of the water column.
Conclusions
We address the question of how pelagial and littoral habitats are integrated into
functional lake ecosystems by asking if a single plant nutrient, phosphorus, can
play the same biomass-limiting role for littoral submerged macrophytes and their
epiphytes, as well as growth of representative consumers, that it does for the
plankton. Enrichments of phosphorus to the environment of natural microbial and
vegetation cornrnunities from a hard water lake demonstrate that phosphorus can
play a concordant growth-regulating role across many aquatic growth-forms.
Epiphytic algae as well as phytoplankton and bacterioplankton were strongly
limited by phosphorus in surnrner. Growth and biomass of submerged macrophytes increased significantly to phosphate released within the calcareous littoral
sediments, but not to inorganic nitrogen. Phosphorus availability did not suppress
growth in the rooted vegetation as severely as it did in the microbial cornrnunities.
Growth rates of two dominant snail species increased rapidly in response to
increased epiphytic algal food supplies under enriched conditions. These results
indicate that phosphorus operates uniformly as a growth-regulating factor among
many growth-forms within the aqueous portions of the ecosystem.
Acknowledgments. We gratefully acknowledge the capable technical assistance of
K. Keagle, J. Sonnad, A.J. Johnson, and S.M. Ford and the critical reviews oftwo
anonymous reviewers. R. Benson of the Sierra Chemical Co. generously provided
324
R.E. Moeller, R.G. Wetzel, and c.w. Osenberg
fertilizer pellets. This work was supported by the National Science Foundation
(DEB-800l190-0l), a NSF pre doctoral fellowship (C.W.O.), and the Department
of Energy (DE-FG05-90ER60930).
References
Barko, J.W.; Smart, R.M. Mobilization of sediment phosphorus by submersed freshwater
macrophytes. Freshwat. Bio!. 10:229-238; 1980.
Bronmark, C. Interactions between epiphytes, macrophytes and freshwater snails: a review.
1. Moll. Stud. 55:299-311; 1989.
Burkholder, 1.A.; Wetzel, R.G. Epiphytic microalgae on a natural substratum in a hardwater
lake: seasonal dynamics of community structure, biomass and ATP content. Arch.
Hydrobiol. Supp!. 83:1-56; 1989.
Burkholder, J.M.; Wetzel, R.G. Alkaline phosphatase and algal biomass on natural and
artificial plants in an oligotrophic lake: re-evaluation of the role of macrophytes as a
phosphorus source for epiphytes. Limno!. Oceanogr. 35:736-747; 1990.
Carignan, R. An empirical model to estimate the relative importance of roots in phosphorus
uptake by aquatic macrophytes. Can. J. Fish. Aquat. Sci. 39:243-247; 1982.
Carignan, R.; Kalff, J. Phosphorus sources for aquatic weeds: water or sediments? Science
207:987-989; 1980.
Carignan, R.; Kalff, J. Phosphorus release by submerged macrophytes: significance to
epiphyton and phytoplankton. Limnol. Oceanogr. 27:419-427; 1982.
Carpenter, S.R.; Kitchell, J.P. Plankton community structure and limnetic primary production. Am. Nat. 124:159-172; 1984.
Cattaneo, A. Grazing on epiphytes. Limno!. Oceanogr. 28:124-132; 1983.
Coveney, M.P.; Wetzel, R.G. Experimental evaluation of conversion factors for the
(3H)thymidine incorporation assay of bacterial secondary productivity. App!. Environ. Microbiol. 54:2018-2026; 1988.
Coveney, M.P.; Wetzel, R.G. Effects of nutrients on specific growth rate ofbacterioplankton
in oligotrophic lake water cultures. App!. Environ. Microbiol. 58: 150-156; 1992.
Coveney, M.P.; Wetzel, R.G. Biomass, production, and specific growth rate of bacterioplankton and coupling to phytoplankton in an oligotrophic lake. Limnol. Oceanogr.
40:1187-1200; 1995.
Cuker, B.E. Competition and coexistence among the grazing snail Lymnaea, chironomidae,
and microcrustacea in an arctic epilithic lacustrine community. Ecology 64: 10-15;
1983.
Dickerman, J.A.; Wetzel, R.G. Clonal growth in Typha lalifolia: population dynamics and
demography of the ramets. J. Eco!. 73:535-552; 1985.
Fairchild, G.W.; Everett, A.C. Effects of nutrient (N, P, C) enrichment upon periphyton
standing crop, species composition and primary production in an oligotrophic softwater
lake. Freshwat. BioI. 19:57-70; 1988.
Fairchild, G.W.; Sherman, J.W. Algal periphyton response to acidity and nutrients in
softwater lakes: lake comparison vs. nutrient enrichment approaches. J. North Am.
Benth. Soc. 12:157-167; 1993.
Grace, J.B.; Wetzel, R.G. Phenotypic and genotypic components of growth and reproduction in Typha talifolia: experimental studies in marshes of differing successional
maturity. Ecology 62:789-801; 1981.
Kalff, 1.; Knoechel, R. Phytoplankton and their dynamics in oligotrophic and eutrophic
lakes. Annu. Rev. Eco!. Syst. 9:475-495; 1978.
Lehman, J.T.; Sandgren, CD. Species-specific rates of growth and grazing loss among
freshwater algae. Limno!. Oceanogr. 30:34-46; 1985.
Lodge, D.M.; Kershner, M.W.; Aloi, J.E.; Covich, A.P. Effects of an omnivorous crayfish
(Orconectes rusticus) on a freshwater littoral food web. Ecology 75: 1265-1281; 1994.
22. Phosphorus Limitation in Lakes
325
Lowe, R.L.; Hunter, R.D. Effect of grazing by Physa integra on periphyton community
structure. 1. North Am. Benth. Soc. 7:29-36; 1988.
Moeller, R.E.; Burkholder, I.M.; Wetzel, R.G. Significance of sedimentary phosphorus to a
submersed freshwater macrophyte (Najas flexilis) and its algal epiphytes. Aquat. Bot.
32:261-281; 1988.
Osenberg, e.W. Body size and the interaction of fish predation and food limitation in a
freshwater snail community. Ph.D. dissertation, Michigan State University, East Lansing; 1988.
Osenberg, C.W. Resource limitation, competition and the influence of life history in a
freshwater snail community. Oecologia 79:512-519; 1989.
Rich, P.H.; Wetzel, R.G.; Thuy, N.V. Distribution, production and role of aquatic macrophytes in a southern Michigan marl lake. Freshwat. BioI. 1:3-21; 1971.
Schindler, D.W. Eutrophication and recovery in experimental lakes: implications for lake
management. Science 184:897-899; 1974.
Schindler, D.W. Factors regulating phytoplankton production and standing crop in the
world's freshwaters. Limnol. Oceanogr. 23:478-486; 1978.
Wetzel, R.G. Longterm dissolved and particulate alkaline phosphatase activity in a hardwater lake in relation to lake stability and phosphorus enrichments. Verh. Int. Verein.
Limnol. 21 :337-349; 1981.
Wetzel, R.G. Limnology. 2nd ed. Philadelphia: WB. Saunders Co.; 1983.
Wetzel, R.G. Land-water interfaces: metabolic and limnological regulators. Verh. Int.
Verein. Limnol. 24:6-24; 1990a.
Wetzel, R.G. Detritus, macrophytes and nutrient cycling in lakes. Mem. 1st. Ital. Idrobiol.
47:233-249; 1990b.
Wetzel, R.G. Microcommunities and microgradients: linking nutrient regeneration, microbial mutualism, and high sustained aquatic primary production. Netherlands 1. Aquat.
Ecol. 27:3-9; 1993.
Wetzel, R.G.; Likens, G.E. Limnological analyses. 2nd ed. New York: Springer-Verlag;
1991.
Wetzel, R.G.; Rich, P.H.; Miller, M.e.; Allen, H.L. Metabolism of dissolved and particulate
detrital carbon in a temperate hard-water lake. Mem. 1st. Ital. Idrobiol. 29 (suppl.):185243; 1972.
Whittaker, R.H. Communities and ecosystems. New York: Macmillan; 1975.
Wright, D.I.; Shapiro, 1. Nutrient reduction by biomanipulation: an unexpected phenomenon and
its possible cause. Verh. Int. Verein. Lirnnol. 22:518-524; 1984.
23. Sources of Organic Carbon in the Food Webs of
Two Florida Lakes Indicated by Stable Isotopes
Mark V. Hoyer, Binhe Gu, and Claire L. Schelske
Introduction
Carbon cycling pathways in lacustrine systems are complex because there are
often multiple sources of organic carbon available to the food webs. Among the
techniques used to delineate carbon flows from organic matter to consumers,
stable isotope analysis may be the most powerful one because isotope compositions of consumers reflect those of the dietary carbon assimilated and incorporated
into their tissues and because no system manipulation is involved. The use of
stable carbon isotopes in food web study is based on these premises: (1) there is a
broad isotope range among different sources of organic matter (Peterson and Fry,
1987) and (2) consumer Dl3 C closely resembles their diets within 1%0 (DeNiro and
Epstein, 1978). Stable isotope analysis has been successfully applied to the investigations of carbon flows in some lacustrine food webs (e.g., Hecky and Hesslein, 1995).
The objective of this study was to determine carbon sources that supported the
food webs of lakes Apopka and Okahumpka, Florida, phytoplankton- and macfOphyte-dominated systems, respectively. We used stable carbon isotopes as natural
tracers to illustrate carbon flows in each food web.
Study Area
Lakes Apopka and Okahumpka are located in central Florida and differ considerably in some major limnological characteristics (Table 23.1). Lake Apopka (28°39'
326
327
23. Organic Carbon in Food Webs
Table 23.1. Typical Limnological Characteristics of Lakes Apopka and
Okahumpka, Florida
Parameter (units)
Apopka
Okahumpka
Secchi disc depth (m)
pH
Dissolved inorganic C (mgIL)
Alkalinity (mglL)
Conductivity (I1S/cm2)
Total nitrogen (mgIL)
Total phosphorus (mgIL)
Chlorophyll a (l1glL)
0.26
9.0
24.7
120
340
4.03
0.192
100.0
Bottom
8.1
26.7
51.3
177
0.95
0.020
5.0
N; 81 °39' W) has a total surface area of 12,400 ha and an average depth of 1.7 m.
This lake is hypereutrophic, with an annual net phytoplankton primary production
300 g C/m 2 (Gale and Reddy, 1994). The water column is highly turbid, as a
combined result of high phytoplankton biomass and frequent resuspension of
the organic sediments. Low light penetration severely restricts the growth of
benthic algae and macrophytes. Lake Okahumpka (28°45'30"; 82°05'02") has
a total surface area of 271 ha and an average depth of 1.1 m. By contrast to
Lake Apopka, Lake Okahumpka is a macrophyte-dominated system with eel
grass (Vallisneria americana) covering more than 90% of the lake surface. As
a result of macrophyte growth, phytoplankton biomass is extremely low. Epiphytic algae are the second important primary producer in this lake (Canfield
and Hoyer, 1992).
Methods
Phytoplankton, epiphytic algae, aquatic macrophytes, invertebrates, and fish were
collected in 1994 and 1995. Owing to low biomass, pure phytoplankton from Lake
Okahumpka and benthic algae from Lake Apopka were not obtainable. Sample
preparations and mass spectrometer determinations for isotope ratios have been
described in detail by Gu et a1. (1996).
13C/12C ratio is expressed in the conventional delta (0) notation, defined as the
per mil (%0) deviation from a standard:
l)X 1,000
The standard is Peedee Belemnite limestone. The analytical precision of these
measurements was 0.1 %0.
328
M.V. Hoyer, B. Gu, and C.L. Schelske
Results and Discussion
The average ol3e of consumers in Lake Apopka ranged from -15 to -8%0,
reflecting carbon source integrated from different phytoplankton signals (-14 to
-3%0) (Fig. 23.1). Aquatic plants and benthic algae are sparse in this lake and were
unlikely to be important to the growth of fish. The small variation in ol3e among
the planktivores (sunfish, gizzard shad, and least killifish) was indicative of
feeding similarity. Blue tilapia is known by its high feeding plasticity, and the
broad range of o13e suggested that there was diet variation within the population
in Lake Apopka. Although a small percentage of the individuals had the cattail
signal (-26%0), most of them derived their carbon source from plankton production. The two benthic feeders (white catfish and brown bullhead) had the ol3e
resembling the phytoplankton signal. Similarly, piscivores also displayed ol3e
signals, reflecting a phytoplankton-based food chain.
The ol3e of the consumers in Lake Okahumpka fell into a narrow range with
one exception. Most of the invertebrates and all fish species (-18 to -16%0)
showed a clear dependence on the epiphytic algal carbon (-19 to -14%0)
(Fig. 23.2). Despite its high abundance, the eelgrass (-8%0) was basically unexploited by consumers. An unidentified species of snail had a ol3e (-12%0) at the
midrange between epiphytic algae and the eelgrass, suggesting that it obtained
013 C (%0)
-30
-20
Bulk Plankton
Microcystis
>0<
Diatoms
Cattail
Sediments
Zooplankton
0
0
t-O-<
t-O---t
c
Grass Shrimp
Redear Sunfish
[J
Bluegill
c
Gizzard Shad
0
Blue Tilapia
o------Q-----I
Least Killifish
0
Brown Bullhead
0
White Catfish
Black Crappie
Largemouth Bass
Florida Gar
0
-10
f{J-I
tOt
I-C-I
t-O-t
0
Figure 23.1. Stable carbon isotope ratios of organic matter and consumers from Lake
Apopka (mean ± ISD).
329
23. Organic Carbon in Food Webs
-20
Eel Grass
Epiphytic Algae
Grass Shrimp
Amphipods
Mayfly Nymph
Snails
Leeches
Bluegill Sunfish
Dollar Sunfish
Redear Sunfish
Lake Chubsucker
Yellow Bullhead
Brown Bullhead
Brook Silverside
Redfin Pickeral
Chain Pickeral
Black Crappie
Largemouth Bass
Florida Gar
-15
-10
-5
[J
>----O----f
0-[}<
>--0---<
[J
tOt
[J
I-Q-4
[J
IoQi
1-0--<
[J
t-O--<
[J
[J
[J
[J
o-[l-i
t[]o
Figure 23.2. Stable carbon isotope ratios of organic matter and consumers from Lake
Okahumpka (mean ± 1SD).
approximately equal amount of its dietary carbon from each source. The large
difference in 8 I3 C between the carnivorous fish and snail indicated that the snail
biomass was not transferred to a higher trophic level.
Our isotope data revealed two major carbon pathways from primary producers
to consumers in the study lakes. Phytoplankton and epiphytic algae were the
energy sources fueling the food webs in lakes Apopka and Okahumpka, respectively. The dependence of consumers on phytoplankton carbon in Lake Apopka is
not surprising because phytoplankton is the only dominant primary producer in
this hypereutrophic lake. Other organic sources, including terrestrial materials,
macrophytes, and benthic algae, are not abundant. Only a few percentages of blue
tilapia showed dependence for growth on cattail detritus.
By contrast, consumers in Lake Okahumpka showed heavy dependence on
epiphytic algal carbon. It is possible that the food web may also derive some
carbon from phytoplankton. However, feeding on the dilute phytoplankton is more
energy expensive than feeding on the abundant epiphytic algae from the macrophyte surface, which are also nutritionally more valuable than phytoplankton
(Hecky et aI., 1993). The most abundant macrophyte, eelgrass, was not used by the
consumers to any great extent. The lack of trophic linkage between the consumers
and aquatic macrophytes in many aquatic systems can be ex.plained by the low
330
M.Y. Hoyer, B. Gu, and c.L. Schelske
nitrogen and high structural carbon contents in aquatic macrophytes (Hecky and
Hesslein, 1995).
In conclusion, our results indicated that stable carbon isotope compositions of
lake biota are excellent indicators of carbon flows in our study lakes. This technique provides insight into the actual carbon sources supporting the growth of
consumers and has the advantage over the conventional methodologies in
elucidating energy pathways in aquatic systems that receive contributions from
different organic sources.
References
Canfield, D.E., Jr.; Hoyer, M.Y. Aquatic macrophytes and their relation to the limnology of
Florida lakes. Florida Department of Natural Resources, Tallahassee, FL; 1992.
DeNiro, M.J.; Epstein, S. Influence of diet on the distribution of carbon isotopes in animals.
Geochim. Cosmochim. Acta 42:495-506; 1978.
Gale, P.M.; Reddy, K.R. Carbon flux between sediment and water column of a shallow,
subtropical, hypereutrophic lake. J. Environ. Qual. 23:965-972; 1994.
Gu, B., Schelske, C.L.; Brenner, M. Relationships between sediment and plankton isotope
ratios (a l3C and a15 N) and primary productivity in Florida lakes. Can. J. Fish. Aquat.
Sci. 53:875-883; 1996.
Hecky, R.E.; Hesslein, R.H. Contributions of benthic algae to lake food webs as revealed by
stable isotope analysis. J. North Am. Benth. Soc. 14:631-653; 1995.
Hecky, R.E., Campbell, P.; Hendzel, L.L. The stoichiometry of carbon, nitrogen, and
phosphorus in particulate matter of lakes and oceans. Limnol. Oceanogr. 38:709-724;
1993.
Peterson, B.J.; Fry, B. Stable isotopes in ecosystem studies. Annu. Rev. Ecol. Syst. 18:293320; 1987.
24. Importance of Physical Structures in Lakes:
The Case of Lake Kinneret and General Implications
Avital Gasith and Sarig Gafny
Introduction
Rock formations, plants, and woody debris are typical sources of physical structure in lakes. They determine physical complexity of littoral habitats, form the
basis for the heterogeneous nature of the nearshore environment, and support
metabolic (organic matter and nutrient dynamic) and nonmetabolic (structure)
related functions (Wetzel and Hough, 1973; Lodge et aI., 1988). Physical structures are often colonized by a diverse assemblage of microorganisms, algae, and
invertebrates and attract predators (mostly fish and macroinvertebrates), which
exploit this rich food resource (e.g., Lodge et aI., 1988; Heck and Crowder,
1991; Diehl, 1993). Structured habitats also provide refugia for prey organisms
(Heck and Crowder, 1991) and are favored as spawning sites (e.g., Goodyear
et aI., 1982; Gafny et aI., 1992). Lake Kinneret (Israel) undergoes relatively
wide water-level fluctuations, providing an opportunity to examine biotic
responses to changes in littoral habitat structure in a relatively large (170 km 2 ),
deep (43 m) lake. Here, we present selected results of our study on the effects
of water-level fluctuation on habitat structure and availability, fish breeding,
and community structure and discuss the importance of physical structures in
lakes of different morphometry.
331
332
A. Gasith and S. Gafny
Results and Discussion
Effect of Water Level on Habitat Structure and Availability
Lake Kinneret water level normally fluctuates within 1.5-2 m. After drought
years, the lake level may fall by 4 m (Gasith and Gafny, 1990). As water levels rise
and fall, large areas along the shoreline are inundated or exposed, changing the
location and structure of the littoral zone. The proportion of shores with rocky
substrate declines from greater than 60% at the highest lake level to less than 10%
as the water level falls by 3.5 m (Gasith and Gafny, 1990). The belt of submerged
rocks also narrows markedly with falling lake level (Fig. 24.1), and stone size
usually becomes smaller. Submerged macrophytes (e.g., Potamogeton pectinatus,
Myriophyllum spicatum) develop sporadically in Lake Kinneret (e.g., Gasith and
Gafny, 1990; Gafny, 1993). Emergent vegetation (mostly Cyperus alopecuroides,
Tamarix jordanensis, Typha angustata, and Phragmites australis) is restricted to
the supralittoral zone during periods of high lake levels. However, dense macrophyte stands develop (up to mean biomass of 1.3 kg dry weight/m 2; total lake shore
biomass of ca. 1,000 metric tons ash-free dry weight) in sandy shores exposed
after lake drawdown (Gasith and Gafny, 1990). Rising lake level inundates the
newly developed vegetation and provides highly structured habitats for a period of
a few months until the plants are uprooted or senesce and decompose (Gasith and
Gafny, 1990; Gafny, 1993). In addition to affecting habitat complexity, water-level
fluctuation markedly changes the availability of substrate colonized by periphyton, which, in the absence of macrophytes, form the main source of organic
matter in the littoral zone (Gafny, 1993).
-208
45
40
35
i' i'
30
..
~
25
"~
20
-209
~
"-0
a;
..
-210 ~
.&!
S
10
5
~
0
OJ
V
...a..
~~
-211 ~
V
15
-212
'o.,.~
-213
FMAMJ J ASONOJ FMAMJ JASON
1991
1992
1_ inundated c:::J Exposed -Water Levell
Figure 24.1. Change of the width of submerged rocky habitats in a selected littoral site
(E-21) of Lake Kinneret during years of low (1991) and high (1992) lake levels.
24. Physical Structures in Lakes
333
Effect of Water-Level Fluctuation on Fish-Habitat Interaction
Of the 24 fish species extant in the lake, 15 may be found during the daytime in the
shallow littoral zone ( m), all at sites with boulders. Eight of these species may also
be found at sites where the substrate is dominated by cobbles, and only four at sites
with structurally simple, sandy substrate (Gafuy, 1993). Most of these fish are
small «80 mm), either small-bodied species or juveniles of larger fish. Under
conditions of high lake level, fish biomass over boulders is often an order of
magnitude higher (80 g/m2 wet weight) than over cobbles, reflecting preference of
more structurally complex habitats over simple ones. During periods of low lake
levels, fish biomass at both habitat types is similar and relatively low «40 glm2 ). This
reflects a decline in fish biomass over boulders and an increase over cobbles. Apparently, under low water-level conditions the fish have no choice but to use any
structured habitat available (Gasith and Goren, 1995). Forcing the fish out of their
preferred habitats can reduce fish survival, partly due to greater predation mortality.
Water-level fluctuation may also affect fish breeding success. During winter
(November-May), large schools of the "lavnun" (Mirogrex terraesanctae, a keystone zooplanktivorous cyprinid) move inshore at night to spawn over rocks in
very shallow waters «50 cm). Only eggs that stick fmnly to the substrate develop
(Gafny et aI., 1992). Winter is also the peak period of epilithon growth in Lake
Kinneret (Gafny, 1993), and algae such as diatoms (e.g., Cymbella sp., Gomphonema sp., Navicula sp.) form a slimy covering on the stones' surface, making the
substrate unsuitable for egg attachment. During periods of low lake levels, the
lavnun uses the "window of opportunity" provided by the rising lake level to
spawn over freshly inundated, temporarily (7-10 days) algae-free rocks found
along the shoreline (Fig. 24.2). A minimal rise in lake level (ca. 30 cm) in the
winter of 1988/89 produced unusual conditions in the littoral zone, and submerged
rocks were overgrown by epilithon throughout most of the littoral zone. We
estimated that more than 90% of the lavnun eggs were lost that winter and
concluded that water-level fluctuation can strongly influence breeding success
and, ultimately, recruitment of young of year (YOY) of the lavnun (Gasith and
Gafuy, 1990; Gafny et al., 1992; Gasith et aI., 1996). This association between
water level, the availability of spawning substrate, and YOY recruitment has been
corroborated by findings from hydroacustic studies of fish abundance in the lake
(Walline et al., 1992; WaHine et aI., 1994).
Water level also influences the availability of preferred spawning substrate for
a blenny (Salaria [Blenniusl jluviatilis), an important zoobenthivorous fish that
spawns on rocks in the littoral zone of Lake Kinneret. Spawn density and size drop
under low lake levels (Aidlin, 1995). However, the same conditions of low lake
levels that deleteriously affected the above species enhance emergent macrophyte
development in exposed shores. When inundated, this vegetation provides
excellent breeding sites for certain cichlids (nest densities >0.8/m2) as well as
refugia and feeding grounds for the larvae and YOY. We can therefore predict
higher recruitment of cichlids in littoral sites around the lake in a rainy year that
follows years of low lake levels.
334
A. Gasith and S. Gafny
400 .-------------------------------------------T1800
_dead eggs
c::J live eggs
~
E 300
u
.
-chlorophyll
~
1200 E
0
0
....
u
g
GI
a.
'"
CI
CI
~
200
>.
GI
r-
'0
0
Z
600
100
e-
o
:E
o
---i----+O
2
7
15
25
30
40
Depth (em)
Figure 24.2. Typical relationship between lavnun (Mirogrex terraesanctae, Cyprinidae)
egg density (mean ± SD) and survival (live and dead eggs), depth, and epilithon biomass
(chlorophyll a) in the littoral zone of Lake Kinneret. (From Gafny et aI., 1992.)
General Implications: The Role of Vegetative Versus Abiotic Formations
and Effect of Lake Morphometry
Most of the knowledge on the structure and function of littoral zones comes from
studies of small vegetated lakes, where the littoral region occupies a relatively
large portion of the lake. In such lakes, both metabolic and structural features of
the littoral zone are important and have been well recognized (e.g., Wetzel and
Hough, 1973; Lodge et aI., 1988). However, the role of the littoral zone in large
deep lakes is less well understood (Danehy et al., 1991; Gasith, 1991). The
availability of vegetated habitats in lakes is influenced by physical (e.g., wave
action, light), water and sediment quality, and biotic controls (e.g., Carpenter and
Lodge, 1986; Gasith and Hoyer, this volume, Chapter 29). In small and shallow
lakes, temporal variability in habitat structure usually follows the seasonal cycle of
macrophyte development. Macrophyte importance in lakes generally declines
with increasing lake size. Duarte et aI. (1986) concluded that, on average, the
percentage of lake area colonized by submerged macrophytes declines with increasing lake size from about 50% in lakes of about 10 ha to about 10% in lakes of
104 ha and to less than 10% in larger lakes. They attributed the relatively higher
importance of emergent vegetation in large lakes to a greater proportion of sheltered bays and floodplains. Large deep lakes often derive physical complexity
from abiotic formations (e.g., Gasith and Gafny, 1990; Danehy et al., 1991;
335
24. Physical Structures in Lakes
c
0
;I
"g
c
/'
."
E
CD
b
C
,....-~
~
a.
a.
~
en
t
0
~
,
,,
,
,
,
~
b
1
....
C1
.....
ia.
High
E
0
u
'0
~
U)
c
,
.!
.5
a1
Low
Small
Lake Size
Figure 24.3. Relationship between lake size and the ratio of supply to demand of littoral
resources and the corresponding intensity of competition over these resources among
consumers that move into the littoral zone (for further details, see text).
Beauchamp et aI., 1994). The size and form of these structures is mostly independent of water-quality conditions but is affected by wave action that sorts for the
larger and heavier particles in the high-energy, nearshore zone and for finer ones in
deeper water (e.g., Walter, 1985). Like plants, the surface of abiotic substrates is often
seasonally modified by growth of periphyton. The availability of littoral habitats with
vegetative or abiotic structure may be altered by water-level fluctuations.
The portion of the lake occupied by the littoral zone is inversely related to basin
slope (Duarte and Kalff, 1986) and to the degree of shoreline regularity, being
smallest in large deep lakes with low shoreline development index. Gasith (1991)
proposed that physical structures and habitat complexity are unique littoral zone
features, and thus littoral resources can be limiting, particularly in large and deep
lakes where the littoral zone occupies a small proportion of the lake area. Similarly, Danehy et ai. (1991) stated that "the relatively few regions of highly structured
habitats [in the Great Lakes] may be more important to the ecology of fish
populations than has been previously recognized." If this is true, we can predict
that in small lakes, where the supply-to-demand ratio of littoral resources is high
(Fig. 24.3, a), the intensity of competitive interactions among consumers that
move from the pelagic zone to use littoral resources should be relatively low (Fig.
24.3, aI). The cross boundary movers are typically adult or large fish that move
between deep and shallow water for feeding, for cover (e.g., at night time; Gasith,
Goren, and Gafny, unpublished data), and seasonally, for breeding. These fish may
use different littoral resources than those used by the "resident," mostly young or
small-bodied littoral species. Higher intensity of competition over littoral resources may be expected in larger deeper lakes (Fig. 24.3, b I), where littoral resources
are relatively limited (Fig. 24.3, b). The underlying assumption is that, other
factors being equal, larger lakes with larger pelagic region can sustain larger
336
A. Gasith and S. Gafny
consumer populations, some of which at times or at some part of their life history
move back and forth between the pelagic and littoral zones. Beyond a certain size
of a water body, the dependency on littoral resources is expected to diminish due
to increased cost of energy expenditure in moving large distances across habitat
boundary, and interhabitat links weaken (Lodge et aI., 1988). For example, most
lacustrine fish use nearshore habitats for spawning and nursery, whereas openwater spawning is usually typical to large deep lakes (Goodyear et aI., 1982) and
is characteristic of ocean-breeding fish. In such a situation, supply of littoral
resources may again exceed the demand (Fig. 24.3, c), lowering the intensity of
competition among cross-boundary movers over these resources (Fig. 24.3, Cl).
Competition for food among resident littoral populations may be as intense in
large and small lakes, depending on the availability of cover, which reduces
predation mortality and increases density-dependent food limitation. High competition for food has been reported in small lakes (Mittelbach, 1988), mostly under
excessive growth of foliated plants species such as milfoil (Heck and Crowder,
1991; Carpenter et al., this volume, Chapter 11).
Conclusions
Water-level fluctuations markedly modify the structure and availability of littoral
zone habitats in Lake Kinneret and can influence fish community structure in a
multiple mechanistic way. During periods of low lake levels, the littoral zone of
Lake Kinneret supports fewer fish. Small fish and YOY are probably most affected
by a reduction in the availability of structured habitats. Low water level is detrimental for fish breeding over rocky substrate but may be beneficial for other
species that exploit the increased availability of vegetated habitats. Two hypotheses on the role of the littoral zone in lakes, and of physical structure in
particular, are evoked from Lake Kinneret study. (1) Structural complexity is
unique to littoral zones making it a potentially limiting factor, particularly in large
deep lakes. In small shallow lakes, the availability oflittoral resources may exceed
the demand placed on them by interhabitat consumers. Thus, intensity of competitive interactions over littoral zone resources is expected to increase with increasing lake size and diminishing proportion of littoral zone area. (2) Under constant
lake levels, abiotic formations may constitute a stable source of structural complexity of littoral habitats, allowing more temporal leeway in resource utilization.
In vegetated lakes, littoral structure and complexity are amenable to biological
feedback and usually follow seasonal plant cycle. The organisms using these
resources are forced to synchronize with the window of opportunity provided by
macrophyte growth. Competitive interactions over vegetative resources is expected to be highest where the period of plant growth is shortest.
In conclusion, the case of Lake Kinneret underscores the importance of water
level fluctuation as a major environmental factor that can strongly influence
habitat structure and related interactions. Due to global climate changes, this
factor may become relevant in lakes that presently exhibit stable water levels.
24. Physical Structures in Lakes
337
Acknowledgments. We thank Larry B. Crowder for constructive remarks and
Merav Bing and Susan Gilman for assistance in manuscript preparation. The study
was supported by the German-Israel Research and Development Fund (BMFrDISUM 00016-GR 00984) and the Kurt Lion Fund for Scientific Cooperation
between the University of Konstanz (Germany) and Tel-Aviv University.
References
Aidlin, M. Biological and ecological aspects of Salariafluviatilis in the littoral zone of Lake
Kinneret. M.Sc. thesis, Tel-Aviv Univ., Israel (Hebrew, English summary); 1995.
Beauchamp, D.A; Byron, E.R.; Wurtsbaug, W.A Summer habitat use by littoral-zone
fishes in Lake Tahoe and the effects of shoreline structures. North Am. 1 Fish. Manage.
14:385-394; 1994.
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Danehy, R.J.; Ringler, N.H.; Gannon, lE. Influence of nearshore structure on growth and
diets of yellow perch (Perea flavescens) and white perch (Marone americana) in
Mexico Bay, Lake Ontario. J. Great Lakes Res. 17:183-193; 1991.
Diehl, S. Effects of habitat structure on resource availability, diet and growth of benthivorous perch, Percafluviatilis. Oikos 67:403-414; 1993.
Duarte, C.M.; Kalff, J. Littoral slope as predictor of the maximum biomass of submerged
macrophyte communities. Limnol. Oceanogr. 31: 1072-1080; 1986.
Duarte, C.M.; Kalff, J.; Peters, R.H. Pattern in biomass and cover of aquatic macrophytes in
lakes. Can. J. Fish. Aquat. Sci. 43: 1900-1908; 1986.
Gafny, S. The effect of substrate type on the structure and function of the littoral zone in
Lake Kinneret, Israel. Ph.D. dissertation. Tel-Aviv Univ., Israel (Hebrew, English summary); 1993.
Gafuy, S.; Gasith, A.; Goren, M. Effect of water level fluctuation on shore spawning of
Mirogrex terraesanctae (Steinitz), (Cyprinidae) in Lake Kinneret, Israel. J. Fish BioI.
41:863-871; 1992.
Gasith, A. Can littoral resources influence ecosystem processes in large, deep lakes? Verh.
Int. Verein. Limnol. 24:1073-1076; 1991.
Gasith, A; Gafny, S. Effects of water level fluctuation on the structure and function of the
littoral zone. In: Tilzer, M.M.; Serruya, c., eds. Large lakes, ecological structure and
function. New York: Springer-Verlag; 1990:156-171.
Gasith, A.; Goren, M. Fish community of the littoral zone of large lakes: the Lake Kinneret
and Lake Conctance experience. Part I-Lake Kinneret. Joint German-Israeli Research
Projects. Final report INCR Tel-Aviv Univ., Israel; 1995.
Gasith, A.; Goren, M.; Gafny, S. Ecological consequences of lowering Lake Kinneret water
level: effect on breeding success of the "Kinneret sardine." In: Steinberger, Y., ed.
Preservation of our world in the wake of change. Jerusalem: ISEEQS Vm:569-573;
1996.
Goodyear, C.D.; Edsall, T.A; Ormsby Dempsey, D.M.; Moss, G.D.; Polanski, P.E. Atlas of
the spawning and nursery areas of Great Lakes fishes. Washington, DC: Fish and
Wildlife Service, U.S. Department of the Interior; 1982.
Heck, K.L.; Crowder, L.B. Habitat structure and predator-prey interactions in vegetated
aquatic systems. In: Bell, S.S.; McCoy, E.D.; Mushinsky, H.R., eds. Habitat structure,
the physical arrangement of objects in space. London: Chapman & Hall; 1991 :281-299.
Lodge, D.M.; Barko, J.W.; Strayer, D.; Melack, J.M.; Mittlebach, G.G.; Howarty, R.W.;
Menge, B.; Titus, J.E. Spatial heterogeneity and habitat interactions in lake communities. In: Carpenter, S.R., ed. Complex interactions in lake communities. New York:
Springer-Verlag; 1988:181-208.
338
A. Gasith and S. Gafuy
Mittelbach, G.G. Competition among refuging sunfishes and effect of fish density on
littoral zone invertebrates. Ecology 68:614-623; 1988.
Walline, P.; Pisanty, S.; Lindem, T. Acoustic assessment of the number of pelagic fish in
Lake Kinneret, Israel. Hydrobiologia 232:153-163; 1992.
Walline, P.; Kalichman, Y.; Ostrovsky, I. Effect of water level fluctuation on the increase of
the "lavnun" population size. In: Zohary, T.; Hambright, K.D., eds. Preliminary assessment of potential impacts of lowering Lake Kinneret water levels to -214 altitude.
Kinneret Limnol. Lab: Special Report T24/94 ILOR (Hebrew); 1994:42-45.
Walter, R.A. Benthic macroinvertebrates. In: Likens, G.E., ed. An ecosystem approach to
aquatic ecology: Mirror Lake and its environment. New York: Springer-Verlag; 1985:
280--288.
Wetzel, R.G.; Hough, R.A. Productivity and the role of aquatic macrophytes. An assessment. Pol. Arch. Hydrobiol. 20:9-19; 1973.
25.
Clear Water Associated with a Dense Chara
Vegetation in the Shallow and Thrbid Lake
Veluwemeer, The Netherlands
Marcel S. Van den Berg, Hugo Coops, Marie-Louise Meijer,
Marten Scheffer, and Jan Simons
Introduction
The presence of submerged aquatic macrophytes in lakes is affected by the
underwater light climate. Lakes with clear water can show abundant macrophyte
vegetation, whereas lakes with turbid water usually have a poor submerged vegetation (Moss, 1990; Scheffer et aI., 1993). Moreover, macrophytes improve their
own light climate by enhancing the water transparency.
Various mechanisms have been proposed to explain the effect of macrophytes
on water transparency. Plants can reduce the concentration of inorganic macronutrients by uptake from the water column, preventing excessive phytoplankton
growth (Ozimek et aI., 1990; Kufel and Ozimek, 1994). FurthelIDore, they may act
as a refugium for zooplankton against predators (Timms and Moss, 1984; Schriver
et aI., 1995). As a consequence, a high grazing pressure inside macrophyte beds
decreases the amount of phytoplanktonic algae. In addition, it has been found in
laboratory studies that macrophytes reduce phytoplankton growth by the release of
allelopathic substances (Wium-Andersen et al., 1982; Hootsmans and Blindow,
1994). Apart from these biological interactions, macrophytes may affect sedimentation and resuspension characteristics of the bottom sediment in favor of clear
water (James and Barko, 1990; Petticrew and Kalff, 1992). In Lake Veluwemeer in
The Netherlands, large areas of clear water have been observed associated with
dense meadows of Chara. These clear areas are in strong contrast to the turbid
339
340
M.S. Van den Berg et al.
nonvegetated parts of the lake (Scheffer et ai., 1994). In 1995, the effect of Chara
on the water transparency in the lake was studied by monitoring some physicochemical and some biological parameters at sites inside and outside the Chara
vegetation.
Study Area
Lake Veluwemeer (3,350 ha; mean depth, 1.45 m) is a large shallow lake in the
center of The Netherlands (Fig. 25.1). The shallow part of the lake is colonized by
charophytes (depth between 30-80 cm). In 1995, Chara vegetation (dominated by
C. aspera) was present in one-third of the lake area, and about 420 ha were covered by
a very dense Chara meadow (maximum biomass, about 500 g DW/m2).
The Netherlands
Transect
Vegetated site
Chara. cover
C:::J 0%
. . 1-15%
_
16-50%
_
51·100%
Figure 25.1. Map of Lake Ve1uwemeer: sample sites and the area covered by Chara in
1995 are indicated.
25. Clear Water Associated with Dense Chara
341
Materials and Methods
Comparison Between Two Sites in 1995
Light attenuation, chlorophyll a, inorganic suspended solids, and detritus were
sampled at two sites in Lake Veluwemeer (Fig. 25.1), an unvegetated and a
Chara-vegetated site at weekly or fortnightly intervals between March-November
1995. The Chara cover at the vegetated site was determined by placing a tube
(surface area, 100 cm2) in the water layer at 10 randomly chosen sites, whereby the
bottom covered area of Chara was estimated by eye. The vertical downward
attenuation coefficient of light (Kd , PAR: 400-700 nm) was measured directly
below the water surface and at 25-cm depth. Water samples (25 L) were taken by
using a tube sampler (perspex tube: length, 1 m; diameter, 5 em). Chlorophyll a
was measured after ethanol extraction. Suspended solids were measured as the
dried (l05°C) fraction remaining after filtering over a 0.45-J.lm membrane filter.
Inorganic suspended solids were measured as the remaining residue after ignition
at 600°C. Detritus concentration was calculated as total suspended solids minus
inorganic suspended solids and algal biomass, assuming that 1 J.lg chlorophyll a
corresponded to 0.07 mg algal biomass (Van Duin, 1992).
Transect Measurements 1995
On July 4, July 18, and August 1, 1995, several physical, chemical, and biological
parameters were sampled along a transect crossing the lake from inside to outside
a dense Chara-vegetated area. Chara cover, suspended solids, chlorophyll a, light
attenuation, and gross sedimentation rate were determined as mentioned above at
nine points along the transect. Sedimentation was measured by placing sediment
traps (PVC; length, 0.5 m; diameter, 0.020 m) 25 cm into the sediment. After 1 or
2 weeks, the content of the traps was collected. The dry weight was determined as
described above. Concentrations of phosphorus (P0 4-P), nitrogen (NOrN, NJLN), and bicarbonate (HC0 3) were measured according to International Standards
(ISO).
On two dates (July 4 and July 18), zooplankton and phytoplankton species
composition as well as their density were determined at seven points along the
transect. To determine phytoplankton composition and density, 1 L of water was
fixed with Lugol's solution. To determine microcrustacean zooplankton composition and density, 25 L water was filtered through a 120-J.lm filter, whereafter the
remaining zooplankton was fixed with 96% ethanol. Biomass of the dominant
zooplankton groups (Daphnia sp. and Bosmina sp.) was estimated by using the
length/weight relationships of Culver et al. (1985). A potential grazing pressure index
(percentage grazed phytoplankton biomass in 1 day [%/day]) was calculated by
assuming that the algal biomass grazed by Daphnia sp. and Bosmina sp. is equal to
their own biomass (Schriver et aI., 1995). On July 4 and 18, the density as well as
the particle sizes (divided in four fractions 1-5 J.lm, 5-30 J.lm, 30-150 J.lm, and
>150 J.lm) was determined by using flow cytometrica1 analysis (Jonker et al.,
1995).
342
M.S. Van den Berg et al.
Analysis of the Light Climate in Lake Veluwemeer
In Lake Veluwemeer, the contribution of humic acids and water to the vertical
attenuation coefficient has been estimated at 0.55/m (Buiteveld, personal communication). The measured attenuation coefficient, chlorophyll a, detritus, and
inorganic suspended solids concentrations were used to estimate the contribution
of these fractions to the light attenuation by multiple regression (Blom et at,
1994), resulting in the model
Kd =0.55 + 0.036 * [inorganic suspended solids] + 0.128 * [detritus] +
0.016
* [chlorophyll a]
where Kd = light attenuation in m-I. Concentrations of inorganic suspended solids
and detritus are in mg/L and the concentration of chlorophyll a in Ilg/L. The model
explained 85% of total variance in K d • The relative contribution of the components
to Kd was calculated from the model.
Results
Comparison Between Two Sites
At the vegetated location, Chara vegetation emerged between April 10 and 25
(Fig. 25.2). Complete cover by the vegetation was reached in July. After August,
the vegetation cover decreased due to grazing by water birds. In November,
shortly before the lake became ice covered, an average cover of 60% was left.
Comparison of the light attenuation coefficients (Kd ) between both sites
showed small differences in March-June (Fig. 25.2). From May to June, the Kd
values at both sites were considerably lower compared with March-April. In
July-August, large differences between the two sites were found: Kd values of the
vegetated site were very low, whereas Kd values of the unvegetated site were high.
In September-November, differences between the sites were small again. The
high transparency at the vegetated location was due to a low contribution of
inorganic suspended solids, detritus, and chlorophyll a. On average, the contribution of inorganic suspended solids, detritus, and chlorophyll a to Kd at the vegetated site in July and August decreased with a factor 46, 6, and 13, respectively,
compared with the unvegetated site.
Transect Measurements
The cover of charophytes inside the meadow was 100%, and at the border of the
meadow, the cover decreased over a short distance to 0%. Along the transect from
outside to inside the vegetation, the water transparency improved with increasing
Chara cover (Fig. 25.3). On July 4 and 18, transparency increased already at a low
cover of Chara. Possibly movement of water from inside to outside the vegetation
caused this phenomenon, because the course of water transparency was closely
:u a.
~ «
~
I
•
Background
~
~
~
c:
0.50 ''It .... I· • I· •
1.50
2.50
•
~
•
....
J:
Q.
ti
o z
~
I-I 0
20
40
rro ]
III
~
"'0
E
......
:a Q.
~ «
l
1., .. I • •
o
::I
~
Detritus
~
c:
I ••
tID
~
Q.
VI
..
oti z~
..,.---------=----------,
Chlorophyll-a
0.50
1.50
2.50
3.50
4.50
80
..........
5.50
100
Inorganic suspended solids
~
tID
• • ••
1·· . . ·1····1 • • I· •
;I' • ••
B. Unvegetated site
Figure 25.2. Seasonal course oflight extinction (Kd ) in Lake Veluwemeer during 1995, and the contribution of background attenuation, inorganic
suspended solids, chlorophyll a, and detritus to the total light attenuation (A) vegetated site and (B) unvegetated site. The course of Chara cover
is indicated by a line.
~
r~j
4.50
5.50
A. Vegetated site
344
M.S. Van den Berg et al.
4 July
18 July
100 _ _ _ _
\ _
~
~
A
50
U
....
2,9
-0.86*
-0.79*
B
1,9
0,9
~
-g
-¥'-----------'
10,--------~~_,
-0.72*
c:
cu
Q.~
c
III~
::l.!....
·2.s
:0",
:::
bj)
6 ;g~
.E
5
0
l~==~~
__.__J
9,----------~
7
-0.69
D
5
3
1--l--------"'-------'
30 , - - - - - - - - - - - - ,
20
-0.86*
-0.71 *
E
10
0+-------,------1
o
200
400
0
200
400
Distance from center of Chara meadow (m)
Figure 25.3. Vegetation cover (A), light attenuation (B), inorganic suspended solids (C),
detritus (D), chlorophyll a (E). bicarbonate (F). phosphate-P (G). ammonium-N (H). and
nitrate-N (I) measured over a transect starting inside and ending outside Chara-vegetated
area on July 4 and 18. 1995. Spearmann correlation coefficients with Chara cover are
given; significant correlations are shown; * P .05. ** P .001.
345
25. Clear Water Associated with Dense Chara
4 July
18 July
1~,-------------~~-.-,
-0.97**
120
80
F
40
O-'-------------------~
+0,14
10
.
O"<r
a..
G
6
2-'-------------------~
0,04
-,-----------------~...,
H
o -'-------------------~
0,6 .,----------------------,
bD
S
z
-0.57
0.4
0,2
o;---------~--------~
o
200
400
0
200
400
Distance from center of Chara meadow (m)
Figure 25.3. (Continued).
related to the course of ions (i.e., Ca and HC0 3) deposited on charophytes. The
improvement of water transparency was related to both lower chlorophyll a,
inorganic suspended solids, and detritus concentrations (Fig. 25.3). Results found
on July 18 were similar to those found on August 1.
Gross sedimentation increased much more along the gradient from vegetation
to the water layer than the suspended matter concentration (Fig. 25.4). Within the
vegetation, the density of large-sized particles decreased more significantly than
small-sized particles (Fig. 25.5). Along the transect, the density of large particles
was very low compared with that of small-sized particles. Phytoplankton (excluding cyanobacteria) showed a comparable pattern, but cyanobacteria showed no
differences in the size distribution between high and low Chara cover. A comparable pattern was found on July 18 (not in figure).
346
M.S. Van den Berg et al.
100 . , - - - - - - - - - - - - - - - - , 1 0 0
......
";"
~
"0
':'
E
10
10
3:
0
......
~
b.O
E
.......
::=9
....
Q)
2n:J
....
~
0
'"0
E
c
N
c
Q)
"0
c:
Q)
Q)
Q.
E
'5
Q)
III
::::l
Vl
III
V>
V>
e
Q
0.1 +-.--,---,--,---,---,---,---,--.,.+ 0.1
o
100
200
300
400
Distance from center of Chara meadow (m)
____ Suspended matter
--+--
Gross sedimentation
Figure 25.4. Suspended matter concentration and gross sedimentation rate over the transect starting inside and ending outside Chara vegetation, three data (July 4, July 18, and
August 3) averaged. Error bars indicate standard error.
On July 18, chlorophyll a concentrations were extremely low at the vegetated
part of the transect (Fig. 25.3D). Also, the relative contribution of cyanobacteria
and green algae to the total density was lower in the vegetation, whereas the share
of flagellates (Cryptomonas sp. and Rhodomonas sp.) was higher (Fig. 25.6A,
Spearman correlation test, P <.05). On July 4, this shift in relative abundances
occurred only in the center of the vegetation. The estimated grazing pressure of
zooplankton on July 4 increased over the transect to the Chara-covered part (Fig.
25.6B, Spearman correlation test, P <.05). On july 18, the grazing pressure tended
to be higher at the border of the Chara meadow. On both dates, the estimated
potential grazing pressure exceeded by at least two times phytoplankton biomass.
o
, .... ---- ___ • ..4
,
,,
~-
..
..
,,.- -IA
--+-
200
1-5j!m
o
200
400
10~'~-----.------~
100 -,
1~
--
5-30 j!m
-- .. -_.
30-150 j!m
Other phytoplankton
10
100
1000
~
> 150 j!m
o
200
400
1~'---------------,
Distance from center of the Chara meadow (m)
400
10~1-------.------­
100
1000
1~
1~0.---------------
Cyanobacteria
Figure 25.5. Densities of suspended particles along the transect with size fractions 1-5 11m, 5-30 11m, 30-150 11m, and >150 11m on
July 4, 1995. The fraction 150 11m of suspended matter was below the detection limit.
~
·in
c:
CI.l
D
a.
:e<d
v
CI.l
VI
c:
.......
E
..........
Suspended matter
A
OC
(j)
<'t!
..c
a.
(j)
>
:p
~
~
0.
.::.:
.s
c
0
u
E
a.
0
'Vi
:p
0
c:
*......
- - -- - -
0
20
100
40-1 _
50
II Others
100
0
II Flagellates
5
~~
~
..._.LJ
4
*
0
20
40
60
80
100
100
100
II Cyanohacteria
@)]]
50
o
5
0
Green algae
18 July 1995
Diatoms
Chara cover along transect (%)
0
_i
-----60ll I I I I I II
----n
80-1 _
...-. 100 ! _
4 July 1995
0
4
::l
Q.
~
III
III
100
1000
10000
100
100
50
5
0
4
100
100
100
Chara cover along transect (%)
0
1000
10000
50
5
o
o
4
1995: (A) relative composition of phytoplankton groups and (B) potential zooplankton grazing pressure.
I!lIiI Potential grazing pressure
Figure 25.6. Composition of plankton over the transect starting inside and ending outside the Chara vegetation on July 4 and IS,
Q..
~
c:
iii ~
+l
&>~
......
c:";.~ ~
btl ......
B
350
M.S. Van den Berg et al.
Ammonium and orthophosphate were not correlated to Chara cover, but on July 4
nitrate and on both dates bicarbonate were strongly negatively correlated with
cover (Fig. 25.3). The pattern of the macronutrient concentrations on July 18 was
similar to August 1.
Discussion
The effect of Chara on biotic and abiotic components of the water layer in Lake
Veluwemeer is pronounced. The clear water above the Chara meadows was
relatively stable, and the border between turbid and clear water was sharp compared with the size of the lake. This suggests that the water exchange between
these two parts was low relative to the rate of the processes that increase water
transparency. As a consequence, during the summer Lake Veluwemeer effectively
consists of two separated parts: one with clear water covered by Chara and a
turbid part characterized by pelagic algae.
The improvement of the light climate inside the vegetation was explained by a
lower concentration of inorganic suspended solids and chlorophyll a and, in less
extent, to lower detritus concentrations. The lower contribution of detritus and
inorganic suspended solids to the light attenuation may be due to the reduction of
resuspension above Chara meadows. The light climate in Lake Veluwemeer is
strongly influenced by wind-induced resuspension (Blom et al., 1994). The dense
charophyte vegetation is likely to restrict the resuspension of bottom sediment
almost completely, whereas sedimentation of particles from the water layer to the
bottom continues. In 1994, it was observed that during high wind velocities the
water above the Characeae became turbid (unpublished data). After stormy
weather conditions (June 21, 1994) the water cleared within I day, indicating that
rapid sedimentation within the Chara meadows may largely explain the difference
in transparency. Moreover, low densities of large-sized particles and a low ratio
between the gross sedimentation rate and suspended matter concentration inside
the vegetation give further support for the hypothesis that the restriction of resuspension is an important mechanism. However, it remains unsolved as to what
extent phytoplankton density was reduced due to sedimentation. Larger algae and
algae without flagella or floating vesicles in particular show net sinking in the
water layer (Reynholds, 1984; Sommer, 1995). The sedimentation of phytoplankton may be increased by higher pH values inside the vegetation, because a
higher rate of calcite incrustation on algae might occur at a high pH (Koschel et aI.,
1983). Indeed, the species composition of phytoplankton shifted to flagellates,
which can escape from sedimentation (Reynholds, 1984). Despite the relatively
high densities of flagellates, however, the absolute density was very low, indicating that mechanisms other than sedimentation are involved in reducing algal
biomass. The decrease of algal densities inside the vegetation may be related to
competition with macrophytes for nutrients, the release of allelopathic substances,
or an increase of zooplankton within the macrophyte stands. In the measurements,
it was observed that during summer no significant difference of dissolved phos-
25. Clear Water Associated with Dense Chara
351
phorus or nitrogen concentration occurred between the site inside and those
outside the Chara vegetation. The uptake of nutrients from the water layer by
rooted macrophytes becomes higher when the ratio between nutrient concentration in the water layer and in the sediment increases (Carignan, 1982; Graneli and
Solander, 1988). Nutrients may be taken up from the water layer by charophytes in
early summer, which was indicated by a decrease in the nitrate concentration on
July 4. On July 18, nitrate and ammonium concentrations along the transect were
very low. Hence, a nitrogen limitation of pelagic algae induced by macrophytes
might occur (Ozimek et aI., 1990).
The toxic effect of allelopathic substances by charophytes has been demonstrated for green algae (Hootsmans and Blindow, 1994) and for cyanobacteria
(Jasser, 1995), but the role of allelopathic substances released by macrophytes
under field conditions is still speculative (Forsberg et al., 1990). The importance of
macrophytes as refuge for zooplankton has often been stressed (Schriver et al.,
1995). Indeed, the grazing pressure of zooplankton was positively correlated to
Chara cover on July 4 and tended to be higher at the border of the charophyte
meadow on July 18. However, the relatively high grazing pressure of zooplankton
outside the vegetation (two to five times the phytoplankton biomass) remains
unexplained but is probably overestimated. The food availability outside the
vegetation may be higher because detritus can be used as a food source for
zooplankton (Jeppesen et al., 1997). Moreover, the relatively high concentration
of small inorganic suspended matter outside the vegetation might negatively affect
the effective grazing rate (Kirk and Gilbert, 1990).
Obviously, reduced resuspension inside dense Chara vegetation explains at
least partly the clearwater patches associated with the charophytes, but other
mechanisms such as nutrient uptake by charophytes and zooplankton grazing may
also be important. To estimate the quantitative contribution of the mechanisms
involved, further experiments are needed.
Acknowledgments. This study was partly financed by Rijkswaterstaat, Directie
IJsselmeergebied. We thank Irmgard Blindow and Christer Bronmark for critically
reviewing the manuscript.
References
Blom, G.; Van Duin, E.H.R.; Verrnaat, J.E. Factors contributing to light attenuation in Lake
Veluwe. In: van Vierssen, W.; Hootsmans, M.J.M; Verrnaat, J. Lake Veluwe, a macrophyte dominated system under eutrophication stress. Dordrecht: Kluwer Academic
Publishers; 1994:158-174.
Carignan, R. An empirical model to estimate the relative importance of roots in phosphorus
uptake by aquatic macrophytes. Can. J. Fish. Aquat. Sci. 39:243-247; 1982.
Culver, D.A.; Boucherle, M.M.; Bean, DJ.; Fletcher, J.W. Biomass of freshwater Crustacean
zooplankton from length weight regressions. Can. J. Fish. Aquat Sci. 42:1380-1390; 1985.
Forsberg, C.; Kleiven, S.; Willen, T. Absence of allelopathic effects of Chara on phytoplankton in situ. Aquat. Bot. 38:289-294; 1990.
Graneli, w.; Solander, D. Influence of aquatic macrophytes on phosphorus cycling in lakes.
Hydrobiologia 170:245-266; 1988.
352
M.S. Van den Berg et al.
Hootsmans, M.J.M.; Blindow, I. Allelopathic limitation of algal growth by macrophytes. In:
van Vierssen, W.; Hootsmans, MJ.M.: Vermaat, J., Lake Veluwe, a macrophyte dominated system under eutrophication stress. Dordrecht: Kluwer Academic Publishers;
1994: 175-192.
James, W.E; Barko, J.W. Macrophyte influence on the zonation of sediment accretion and
composition in a north-temperate reservoir. Arch. Hydrobio!. 120:129-142; 1990.
Jasser, I. The influence of macrophytes on a phytoplankton community in experimental
conditions. Hydrobiologia 306:21-32; 1995.
Jeppesen, E.; Jensen, J.P.; S~mdergaard, M.; Lauridsen, T.; Pedersen, LJ.; Jensen, L. Top
down control in freshwater lakes: the role of nutrient state, submerged macrophytes and
water depth. Hydrobiologia 342/343:151-164; 1997.
Jonker, R.R.; Meulemans, J.T.; Dubelaar, G.B.J.; Wilkins, M.E; Ringelberg, J. Flowcytometry: a powerful tool in analysis of biomass distributions in phytoplankton. Wat.
Sci. Techn. 32:177-182; 1995.
Kirk, K.L.; Gilbert, J.J. Suspended clay and the population dynamics of planktonic rotifers
and Cladocerans. Ecology 71:1741-1755; 1990.
Koschel, R.; Benndorf, 1.; Proft, G.; Recknagel, E Calcite precipitation as a natural control
mechanism of eutrophication. Arch. Hydrobio!. 98:380-408; 1983.
Kufel, L.; Ozimek, T. Can Chara control phosphorus cycling in Lake Luknajno (Poland)?
Hydrobiologia 2751276:277-283; 1994.
Moss, B. Engineering and biological approaches to the restoration from eutrophication in
which aquatic plant communities are important components. Hydrobiologia 275/276:
367-377; 1990.
Ozimek, T.; Gulati, R.D.; van Donk, E. Can macrophytes be useful in biomanipulation of
lakes, the Lake Zwemlust example. Hydrobiologia 200/201 :399-407; 1990.
Peuicrew, E.L.; Kalff, 1. Water flow and clay retention in submerged macrophyte beds. Can.
J. Fish. Aquat. Sci. 49:2483-2489; 1992.
Reynholds, C.S. The ecology of freshwater phytoplankton. London: Cambridge University
Press; 1984.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Eco!. Evo!. 8:276-279; 1993.
Scheffer, M.; Van den Berg, M.; Breukelaar, A.; Breukers, C.; Coops, H.; Doef, R.; Meijer,
M.-L. Vegetated areas with clear water in turbid shallow lakes. Aquat. Bot. 193-196;
1994.
Schriver, P.; Bjjgestrand, J.; Jeppesen, E.; Sjjndergaard, M. Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large scale enclosure experiments in a shallow eutrophic lake. Freshwat. BioI. 33:255-270; 1995.
Sommer, U. Planktologie. Berlin: Springer-Verlag; 1995.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limno!. Oceanogr. 29:472-486; 1984.
Van Duin, E.H.S. Sediment transport, light and algal growth in the Markermeer. Thesis,
Agriculture University, Wageningen; 1992.
Wium-Andersen, S.; Anthoni, U.c.; Christophersen, C.; Houen, G. Allelopathic effects on
phytoplankton by substances isolated from aquatic macrophytes (CharaIes). Oikos 39:
187-190; 1982.
26. Alternative Stable States in Shallow
Lakes: What Causes a Shift?
Irmgard Blindow, Anders Hargeby, and Gunnar Andersson
Introduction
Lake Takern and Lake Krankesjon, two shallow, moderately eutrophic, calciumrich lakes in southern Sweden have shifted between turbid and clearwater states
several times during the past decades (Fig. 26.1). Lake Krankesjon shifted from a
clearwater state with abundant submerged vegetation to a turbid state with sparse
vegetation during the mid-1970s (Karlsson et aI., 1976) and back to a clearwater
state during 1985. Today, the lake is in the clearwater state, with abundant submerged vegetation dominated by Charophyta (Blindow et aI., 1993; Fig. 26.2).
Both shifts coincided with deviations from the average water level. During the
mid-1970s, the water level during spring and summer was about 15 cm higher than
average, whereas it was about 10 cm lower than average during 1983-1985
(Blindow, 1992).
In Lake Takern, the submerged vegetation disappeared at least twice during the
beginning of the century due to catastrophic events (extremely low water level
causing dry-out during summer and damage by ice during winter, respectively),
but it soon recovered. During the 1950s, a similar disappearance of submerged
plants occurred due to dry-out, causing a shift to a turbid state. In the mid-1960s,
the lake shifted back to a clearwater state with abundant submerged vegetation
dominated by Charophyta. This shift took place after the application of a new
water regime with lower amplitudes in water level. Since then, the lake has been
in a clearwater state (Ekstam, 1975; Blindow et al., 1993). However, during the
353
354
I. Blindow, A. Hargeby, and G. Andersson
....----------------.
--------~------.==----------
L.T~em
•
1890
1900
1910
1920
1930
1940
1950
L. Krankesjoo
1960
1970
1980
1990
2000
Figure 26.1. Long-tenn shifts between clearwater (thin lines) and turbid (thick lines) states
in Lake Takern and Lake Krankesjon, schematically.
1600
1400
1200
.
<"I
e
1000
800
600
400
200
0
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Figure 26.2. Average biomass (fresh weight) of submerged macrophytes per lake surface
outside the reed belts of Lake Takem (black symbols) and Lake Krankesjon (white symbols). Biomass values were calculated from data on distribution of submerged macrophytes
(obtained by investigation from boat, combined with air photographs during several years)
multiplied with biomass values for different macrophyte species (sampled with Plexiglas
corer). Values for 1995 and 1996 plant distribution are rough estimates as the investigation
of plant distribution was hampered by high water turbidity.
26. Stable States in Shallow Lakes
355
past two summers the vegetation developed slowly, simultaneously with extended
periods of turbid water. The monitoring undertaken during these years may give
information regarding shifts between clearwater and turbid states.
Recent Observations in Lake Takern
During 1994-1996, the development of submerged vegetation was monitored in
Lake T3kern during spring and summer. Samples for biomass of submerged
vegetation were taken about three times per month with a Plexiglas core sampler
in a stand of Chara tomentosa. This species is one of the dominant submerged
macrophytes in Lake Takern and hibernates as a green plant (Blindow, 1992). The
samples were frozen, dried at 105°C (24 hours) and 550T (2 hours) for determination of dry weight and ash-free dry weight, respectively. Whole-column water
samples were taken with a Plexiglas core in an unvegetated area close to the site
where the plant samples were taken. Apart of the sample was frozen and analyzed
later for suspended material (GF/C-filtered sample). For analysis of chlorophyll.,
water was filtered immediately (GF/C). The filters were deep frozen for later
analysis according to standard procedures by using extraction with methanol.
During 1994, the biomass of Chara tomentosa decreased during the end of
April and increased by the end of May. Simultaneously with this increase, water
turbidity decreased (Fig. 26.3). During the summer, the plants reached the water
surface, and estimated values for overall biomass of submerged macrophytes
(including Chara tomentosa) were similar to values obtained during previous
years (Fig. 26.2).
Also during 1995, the biomass of Chara tomentosa decreased by the end of
April. In the opposite to 1994, however, only a minor increase of plant biomass
was observed during spring and summer, and the water was turbid throughout the
whole period of investigation (Fig. 26.3). In contrast to all previous years when the
submerged vegetation has been investigated (yearly since 1983, except 1993),
neither Chara tomentosa nor any other species of submerged plants reached the
water surface during summer 1995. Consistently, estimated biomass for submerged macrophytes was lower than in previous years (Fig. 26.2). First during the
autumn, submerged plants reached water surface in limited areas of the lake (L.
Gezelius, personal communication). Values for chlorophyll taken during the summer of 1995 were almost twice as high as the years before (Fig. 26.4).
During 1996, the biomass of Chara tomentosa was low at the beginning of
the season and remained low throughout the summer, with a minor increase
during the autumn. The turbidity was low during the beginning of the season
but increased continuously (Fig. 26.3). In the opposite to 1994 and 1995, the
vegetation grew patchily, and most of the areas previously covered with dense
stands of Chara tomentosa were vegetation-free. Estimates in July indicated
that the total biomass of submerged macrophytes was even lower than during
1995 (Fig. 26.2). Values for summer chlorophyll increased further compared
with 1995 (Fig. 26.4).
356
I. Blindow, A. Hargeby, and G. Andersson
250
200
N
E
:g
'"'"
III
E
150
0
iIi
100
70
f
r
50
40
••
f~n'
50
60
•
E
:g
''""
III
E
III
D0>
'0
c
Co
•
20
•
June
••
•
••
••
150
July
en'"
::J
10
August
•
•
September
70
60
'§,
••
50
100
.s
'"
Q)
40
•
0
iIi
t
0>
250
&'
'"0>
13
0
May
April
200
.s
'0
30
0
1994
'§,
•
13
tIII
a..
'0
30
~
c
0>
Co
20
50
QO
££ £0 0 O~
?
~
0
1995
en'"
:J
10
0
April
May
June
July
August
September
Figure 26.3. Biomass of submerged vegetation (Cham tomentosa; white circles,
mean ± standard error) and turbidity (black circles) in Lake Takern during 1994 (above),
1995 (middle), and 1996 (next page).
Summer densities of Cladocera (including Daphnia) were low during all 3
years. Spring peak densities of both total number of Cladocera and Daphnia were
lower during 1994 than in the years of turbid water (1995 and 1996; Table
26.1). The few data that exist on fish assemblage are hard to evaluate in terms of
biomass for the whole lake. Catches with survey nets in July in a restricted area
357
26. Stable States in Shallow Lakes
250
•
200
l'
S!
150
III
III
IG
E
•
0
in
100
••
• •
• •
•
•
60
~
S
I/)
~
40
2
2 ~ ~ 222 2 Y 2
f
.2
t::
IG
a..
"C
30
CD
"C
C.
CD
Co
20
I/)
:::I
en
10
0
0
1996
70
50
~
50
•
•
April
May
June
July
August September
Figure 26.3. (Continued).
-30
T
25
...
T
20
0
~
'-'
:::l
~
i
15
T
10
5
o
I
1986
I
1987
1991
1992
1994
1995
1996
Figure 26.4. Chlorophyll concentration in Lake Takem during different years. Mean
values of samples taken from June to August (only single samples taken during 19861992). Standard error is given for samples taken during 1994 to 1996.
1. Blindow, A. Hargeby, and G. Andersson
358
Table 26.1. Numbers of Cladocera and Daphnia During Spring (Peak Values) and
Summer (Average Values for Samples Taken During July and August) in Lake Takem
Year
Spring peak
numbers, all
Cladocera"
Spring peak
numbers,
Daphnia Spp.b
1994
1995
1996
62
304
430
16
136
54
Summer
a verage, all
Cladocera
II (n
II (n
13 (n
Summer
average,
Daphnia
o
o
= 2)
= 5)
= 5)
2
"Dominant species: Bosmina iongirostris.
bOominant species: Daphnia iongispina.
(about 2 km2 ) in the center of the lake, however, do not indicate drastic changes in
abundance or species composition (Table 26.2) between 1990 and 1996.
Discussion
Shifts between the turbid and the clearwater state in shallow lakes are often caused
by human interference. Increased nutrient loading to shallow lakes has in many
cases led to disappearance of submerged vegetation and a drastic decrease of water
transparency (e.g., Scheffer et aI., 1994), whereas reduction of the fish stock has
been applied in several cases to attain a switch back to the clearwater state (e.g.,
van Donk et al., 1990; Meijer et aI., 1994). Similar to Lake Tomahawk Lagoon in
New Zealand (Mitchell et aI., 1988; Mitchell 1989), both Lake Takem and Lake
Krankesjon several times switched "spontaneously" between the two states, without any obvious influence or manipulation from "outside." Furthermore, the
higher densities of Cladocera (including Daphnia) during 1995 and 1996 compared with 1994 suggest that the observed change in Lake Takern was not caused
by top-down mechanisms. Instead, the recent results from Lake Takem presented
above support our earlier suggestion (Blindow et aI., 1993) that water-level fluc-
Table 26.2. Fish Catches (kg) in Survey Nets (4-1 DO-mm Mesh) in an Area with Sparse Vegetation
(Myriophyllum spicatum in 1985 and 1986, Sparsely Occurring and Low-Grown Chara tomentosa
1990 and 1996)
1985
1986
1990
1996
a
n
Pike"
Roach a
Rudd"
Tencha
Crucian
carp"
Perch a
Ruffe"
2
3
3
5
l.l ± 1.6
2.4 ±0.5
1.5 ± 0.5
0.6 ±0.2
1.8 ± 0.5
0.1 ±0.1
0.2 ±0.2
0.1 ±0.2
0.01 ±0.03
3.7±0.7
2.8 ± 1.8
1.2 ± 1.5
1.5 ± 1.2
0.6±0.2
0.2 ± 0.4
0.0
0.0
2.1 ± 0.2
0.4 ± 0.2
0.5 ± 0.4
0.5 ± 0.2
2.1 ± 1.2
0.01 ±0.01
0.01 ±0.01
0.02±0.03
Means ±SO.
0.3 ±0.5
0.5 ±0.8
0.9 ± 2.0
359
26. Stable States in Shallow Lakes
450
440
430
420
410
400
1985
1995
1994
390
380
370~'--,,--.--.--.--.--r--r--.---'--.---.-
Figure 26.5. Water level (monthly average of daily readings) in Lake Takem during 1985
(white squares), 1994 (black circles), and 1995 (black squares). Data obtained from the
Swedish Meteorological and Hydrological Institute. Dotted line: water level aimed at
according to court decision.
tuations are an important factor affecting the submerged vegetation and eventually
causing a switch. High spring water level in Lake Takem during 1985 was the
most probable cause of the relatively low biomass of submerged vegetation
registered in the summer of that year (Fig. 26.5). However, this disturbance was
not sufficient to cause a switch to the turbid state (Blindow, 1992). Also during
1995, the spring water level was relatively high (Fig. 26.5). Compared with 1985,
the deviation from normal water level was lower but occurred during a longer time
period and coincided with low temperature during April to mid-May (unpublished
data). High precipitation during this period probably caused a somewhat higher
nutrient loading to the lake due to runoff from the agricultural area surrounding the
lake. This is indicated by the fact that both water turbidity and chlorophyll
concentrations at the beginning of the growing season were higher during 1995
than during 1994. We suggest that the combination of these factors-high spring
water level, high turbidity, and low temperature-was the reason for the substantial decline in submerged macrophytes during 1995.
Despite lower turbidity in the beginning of the growing season, the transition
toward a turbid state continued during 1996 in Lake Tiikem. The reason for this
development may be the low spring temperature also that year, in combination
with reduced biomass of hibernating Chara tomentosa and possibly other submerged macrophytes. Thus, the unfavorable conditions in 1995 may have affected
the development of submerged plants the following year, weakening the feedback
mechanisms that stabilize the clearwater stable state (Scheffer et al., 1993).
360
I. Blindow, A. Hargeby, and G. Andersson
Acknowledgment. This study was financially supported by the World Wide Fund
of Nature, Sweden ("Tiikernfonden").
References
Blindow, 1. Long- and short-term dynamics of submerged macrophytes in two shallow
eutrophic lakes. Freshwat. BioI. 28: 15-27; 1992.
Blindow, I.; Andersson, G.; Hargeby, A.; Johansson, S. Long-term pattern of alternative
stable states in two shallow eutrophic lakes. Freshwat. BioI. 30: 159-167; 1993.
Ekstam, U. Foriindringar av fagelfauna och miljo i och vid Takern 1850-1974 (Changes of
the avifauna and the nature and environment of Lake Takern in 1850-1974). Var
Fagelvarld 34:268-282 (in Swedish with English summary); 1975.
Karlsson, J.; Lindgren, A.; Rudebeck, G. Drastiska fOrandringar i vegetation och fagelfauna
i Krankesjon och Bjorkesakrasjon 1973-1976 (Drastic changes in vegetation and bird
fauna in Lake Krankesjon and Lake Bjorkesakrasjon, South Sweden, in 1973-1976).
Anser 15: 165-184 (in Swedish with English summary); 1976.
Meijer, M.-L.; van Nes, E.H.; Lammens, E.H.R.R.; Gu1ati, RD.; Grimm, M.P.; Backx, J.;
Hollebeek, P.; Blaauw, E.M.; Breukelaar, A.W. The consequences of a drastic fish stock
reduction in the large and shallow Lake Wo1derwijd, The Netherlands. Can we understand what happened? Hydrobiologia 275/276:31--42; 1994.
Mitchell, S.F. Primary production in a shallow eutrophic lake dominated alternately by
phytoplankton and by submerged macrophytes. Aquat. Bot. 33:101-110; 1989.
Mitchell, S.F.; Hamilton, D.P.; MacGibbon, W.S.; Nayar, P.K.B.; Reynolds, R.N. Interrelations between phytoplankton, submerged macrophytes, black swans (Cygnus atralus)
and zooplankton in a shallow New Zealand lake. Int. Rev. Ges. Hydrobio\' 73:145-170;
1988.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evo\. 8:275-279; 1993.
Scheffer, M.; van den Berg, M.; Breukelaar, A.; Breukers, c.; Coops, H.; Doef, R.; Meijer,
M.-L. Vegetated areas with clear water in turbid shallow lakes. Aquat. Bot. 49: 193-196;
1994.
van Donk, E.; Grimm, M.P.; Gulati, R.D.; Klein Breteler, J.P.G. Whole-lake food web
manipulation as a means to study community interactions in a small ecosystem. Hydrobiologia 200/201 :275-291; 1990.
27.
Clear and Thrbid Water in Shallow Norwegian
Lakes Related to Submerged Vegetation
Bj0rn A. Faafeng and Marit Mjelde
Introduction
Timms and Moss (1984) suggested that fertile shallow lakes may have alternative
stable states, a clearwater state with dense vegetation and a turbid water state
dominated by phytoplankton and with little submerged and floating-leaved vegetation. This phenomenon has also been observed and discussed by several other authors
(e.g., Irvine et al., 1990; Jeppesen et al., 1990; van Donk et al., 1990; Blindow et al.,
1993; Scheffer et al., 1993). According to the model of Scheller (1989, 1990), the
main controlling factor for the two alternati ve states is the turbidity of water regulating
the vertical light penetration. When the nutrient level increases, phytoplankton growth
is often stimulated, which in turn increases the turbidity. This leads to increased light
attenuation and a reduction of the maximum growth depth of submerged vegetation.
Increased nutrient concentrations will therefore reduce the bolttom area covered by
submerged vegetation until it is virtually absent. In shallow lakes with most bottom
areas at similar depths, the change from high plant cover to plantless bottoms may be
abrupt. Uncovered sediments in shallow lakes are much more vulnerable to resuspension and more easily give rise to turbid water during periods of wind and wave action
than in lakes with plant-covered sediments. Benthivorous fish may be favored under
these circumstances and add to the turbidity by foraging on the sediment. A more-orless stable turbid state may be the result.
On the contrary, when dense submerged vegetation covers a large part of the
bottom area, these plants may successfully remain with sparse development of
361
362
B.A. Faafeng and M. MjeJde
phytoplankton even at high nutrient concentrations (Timms and Moss, 1984; van
Donk et al., 1993; Mjelde and Faafeng, 1997). The water adjacent to these
macrophyte stands may be clear, also inside patches of vegetation (Scheffer et al.,
1994). Several mechanisms connected with the submerged macrophytes support the
stability of the clearwater stage: competition with phytoplankton for available
nutrients (van Donk et al., 1993) and with epiphytes for light (Phillips et al., 1978;
Sand-Jensen and S!lindergaard, 1981), release of allelopathic substances (WiumAndersen, 1987), and their acting as refuge for Daphnia and piscivorous fish
(Timms and Moss, 1984; Grimm, 1989; Schriver et al., 1995). These mechanisms
add to the turbidity effect.
In oligotrophic lakes, the two alternative stable states will not occur, as the phytoplankton biomass is not sufficiently high. Jeppesen et al. (1990) observed the two
alternative states in Danish lakes with a total P concentration between approximately 50
and 125 Ilg total PIL, and the clearwater state has also been observed in very small lakes
at much higher concentrations (Balls et al., 1985; Jeppesen et al., 1990).
The transition between a "clear" and a "turbid" state is so far, however, highly
subjective. In lakes with a total P concentration of 20 Ilg PIL, the average concentration of chlorophyll (Chla) throughout the growth season may vary at least
between 5 and 20 Ilg ChlaIL (Faafeng and Hessen, 1993). It is obvious that in lakes
with a total P of 20 Ilg PIL, the growth potential is normally much lower and the
expected minimum transparency caused by phytoplankton much higher than at
total P levels of, for instance, 200-300 Ilg PIL. Therefore, although the low light
intensity needed to shade out the macrophytes is the same, we suggest that in this
respect the limit between clear and turbid should relate to their total P levels. In a
previous paper (Mjelde and Faafeng, 1997) we demonstrated that the clearwater
state appeared in the investigated lakes with mean depths less than 1.9 m and when
the bottom area covered with vegetation was greater than 50%. We also suggested
that the average Chlaltotal P ratio over the growth season (May-September)
indicates whether the water is clear (close to 1:10) or turbid (close to 1:1). The
focus of this chapter is to give a more detailed study of the clear and turbid states
over high and low total P concentrations by using the ratios Chlaltotal P and total
P/transparency on a data set from shallow Norwegian lakes.
Materials and Methods
This study includes 10 small «1 km2) and shallow (mean depth, <4 m) lakes with
an average total P greater than 20 Ilg PIL. The lakes are situated in northern and
southern parts of Norway and cover latitudes between 58-69°N. All lakes were
shallow enough to allow growth of vegetation over a major part of the bottom area
under favorable light conditions. Integrated water samples were taken from the
phototrophic layer (twice the Secchi depth or at most down to 0.5 m above the lake
bottom) at least four times during May-September for analysis of water chemistry
and quantitative phytoplankton and zooplankton. Total P was analyzed with a
Technicon autoanalyzer after persulfate digestion, and chlorophyll a was measured
363
27. Clear and Turbid Water in Shallow Norwegian Lakes
spectrophotometric ally after acetone extraction. Between 1988-1993, at least one
growth season was studied in each lake.
The distribution of rooted submerged vegetation toward depth in lakes is
normally limited by light, preventing vegetation cover in deeper areas (Vant et al.,
1986). When lakes are shallow and clear, submerged and floating-leaved plants
may cover the whole bottom area. In our survey, we used Secchi disc transparency
as a measure for the light conditions. Transparency is often a realistic alternative
to measurement of vertical light attenuation and is a useful approximation (Canfield et al., 1985; Chambers and Kalff, 1985).
Major phytoplankton blooms may occur in the spring (May) before the submerged vegetation has established, and this may seriously affect the average
growth-season values of nutrients, chlorophyll, and transparency. This phenomenon was especially prominent in some of the lakes with conspicuous unrooted
growth of Ceratophyllum demersum from early June. To avoid this problem, we
therefore use "average late summer" values (ALS) calculated as an average of the
values from July, August, and September.
The submerged vegetation was studied once in each lake (in late July-September) during the years 1992-1996 by using a hydroscope and by dredging from a
boat. The species distribution was recorded and the bottom cover estimated. The
methods are discussed in further detail in Mjelde and Faafeng (1997).
Results
The lakes had ALS total P concentrations ranging from 23 to more than 500 Ilg P/L
(Table 27.1), whereas average transparency ranged between 0.3-3.8 m. In three of
Table 27.1. Late Summer Average V alu es (J ul y-Septem ber) of Ph osph oru s, Chlorophyll a, and Secchi Disc Transparency for Each Year in the 10 Lakes
Lake
Year
Total P (/-tg PIL)
ChI. (/-tglL)
Transparency (m)
S~ylandsvatnet
1992
1988
1992
1993
1988
1992
1993
1993
1992
1992
1992
1992
1992
1993
594
223
228
307
188
206
165
100
86
61
52
30
27
23
23
82
94
109
139
111
79
43
13
17
16
3
7
3
1.7a
0stensj~vann
0stensj~vann
0stensj~vann
Hellesj~vann
Hellesj~vann
Hellesj~vann
Lille Gleinsvatn
Smokkevatn
Kringelvatn
Mosvatn
Altervatn
Haversvatn
Stavsengvatn
aEstimated values (see text).
1.0
0.6
0.6
0.3
0.3
0.5
0.9
2.0
1.9a
1.7
3.3 a
2.4
3.8
364
B.A. Faafeng and M. Mjelde
Table 27.2. Lake Morphometry and Bottom Cover of Submerged and
Floating-Leaved Vegetation
Lake
Sj3ylandsvatn
0stensjj3vann
Hellesjj3vann
Lille Gleinsvatn
Smokkevatn
Kringelvatn
Mosvatn
Altervatn
Haversvatn
Stavsengvatn
Lake area
(km2)
Mean depth
(m)
Bottom cover
0.65
0.31
0.53
0.10
0.14
0.09
0.50
0.08
0.15
0.06
O.4a
1.9
1.3
4.0
1.2
1.1
100
20
32
28
74
74
1.6
O.5a
2.4
1.9
100
17
61
(%)
64
aEstimated values.
.......... - ................................. __ ................................................._-_ .........................._;;... ....;
1000
1:1
..........
a
its
oS!
100
........
........
o
10
--- --10
o
o
_--0-----
---
1000
Figure 27.1. Total P versus chlorophyll a in the 10 lakes (late summer average values)_
Lakes with high bottom cover of vegetation are shown as white symbols whereas lakes with
low bottom cover are shown with black symbols_ I: I and 1: 10 lines are indicated_ Connected points are several years from the same lake_ All lakes with high plant cover are
perceived as "clear," whereas lakes with low plant cover, except one (Haversvatn: lower left
black symbol), are "turbid."
27. Clear and Turbid Water in Shallow Norwegian Lakes
365
10,··················································· ......................................................................................................................................................,
o
J;.
• veg. <50%
o veg. >50%
J;. veg. >50% est.
0.1+-----~----~~~~~~-r------~--~-r--~~~rl
10
1000
Figure 27.2. Concentration of total P versus transparency in lakes with high bottom cover
of vegetation (white symbols) and with low bottom cover (black symbols). In lakes with
transparency higher than the actual maximum depth and with high bottom cover, transparency has been estimated (white triangles). All values are late summer averages. Connected points are several years from the same lake. A linear regression line from 76
lake-years from deep and shallow Norwegian lakes is included (see text).
the lakes, one or more of the Secchi disc readings were higher than the maximum
depth of the lake, or the vegetation was too dense to allow proper transparency
measurements. For these lakes, transparency was estimated from the Chla versus
transparency linear regression line of the other seven lakes covering 11 lake-years:
(transparency = 4.225 - 0.813*ln(Chla), ,-2 = 0.947, P <.001). We found no
particular reason to expect a severe deviation from this line for the three lakes
in question. In 6 of the 10 lakes, the submerged vegetation covered 50% of the
lake bottom area or more (Table 27.2), and in 2 of the lakes, the whole bottom
area was covered with vegetation, Ceratophyllum demersum being the dominant species.
In a plot of average growth season total P versus ChIa, we previously found that
most lakes with greater than 50% bottom cover of vegetation had a ChIaltotal P
ratio close to 1: 10, whereas most lakes with less than 50% cover had a Chlaltotal P
ratio greater than close to 1:1 (Mjelde and Faafeng, 1997; Fig. 27.1). By using
ALS values instead of "growth season averages," a better distinction between the
lakes with high versus low cover of submerged vegetation was found.
The total P versus transparency plot (Fig. 27.2) shows a clear trend toward
higher transparency in lakes with high vegetation cover than in lakes with a lower
cover, both at high and low P concentrations. In this figure, a regression line from
76 lake-years is also given from both deep and shallow Norwegian lakes with total
366
B.A. Paafeng and M. Mje\de
P concentrations ranging between 20-200 !lg PIL; In(transparency) = 3.306 0.770*ln(total P), r = 0.60, P <.01 (Faafeng, unpubl.).
Discussion
The shallow lakes with a high bottom cover of vegetation were obviously among
those with the highest transparency at a certain total P concentration compared
with other deep and shallow lakes as indicated by the regression line (Fig. 27.2).
This is due to a lower Chla concentration as shown in Figure 27.1. This fact is even
more prominent when taking into account the general trend for shallow unstratified lakes to have a higher phytoplankton yield per unit total P than deeper
lakes (Riley and Prepas, 1985; Mazumder, 1994).
Unfortunately, we have no shallow lakes in our investigation representing the
turbid state in the lower end of the total P range. Lake Haversvatn, with a
submerged bottom cover of only 17% and represented by a black symbol in the
figures, does not produce the same phytoplankton yield as might be expected from
its total P concentration (Faafeng and Hessen, 1993), and it is consequently not
classified as turbid. Concentrations less than detection limits of available inorganic nitrogen and a constantly small population of Daphnia «11 individualsIL)
during May-September support this discrepancy and indicate N limitation. The
lake with the highest total P concentration, Lake S0ylandsvatn, is probably also N
limited throughout the growth season (Mjelde and Faafeng, 1997) and therefore
deviates considerably from the general trend of both Figures 27.1 and 27.2. It is a
drawback to our analysis that we have related our discussion to total P only.
However, at lower total P concentrations, which prevail in Norwegian lakes, total
P is often a fair approximation of the growth potential of the phytoplankton.
Timms and Moss (1984) referred to "very clear water «20!lg ChlaIL)" in some
lakes in the Norfolk Broads with dense plant communities "despite extremely high
concentrations of phosphate and nitrogen." This fits well with the observations of
van Donk and Gulati (1995), who distinguish between the "clear" and the "turbid"
states in Lake Zwemlust at average May-September chlorophyll concentrations
less than or greater than 30 !lg ChlaiL. Lake Zwem1ust had very high average total
P concentrations ranging from 1,863 to 2,697 !lg total PIL in the years 1986, 1989,
and 1992, and the lake phytoplankton was obviously not P limited during the
investigated years. In years with dense submerged vegetation, most of the N was
stored in the macrophytes and the phytoplankton became N limited. Their Table 1
shows that the corresponding average transparency of clear and turbid states was
greater than 1.9 m or less than 1.5 ill, respectively. All the "clear" lakes in our
survey also had ALS chlorophyll concentrations less than 20-30 !lg ChlaIL (Fig.
27.1) and a transparency greater than 1.5-2 m (Fig. 27.2). This fits well also with
Hargeby et al. (1994), who observed "turbid" water in Lake Krankesjon. During
lune-September, the average Chla concentrations in Krankesjon ranged, for example, between 25-30 !lg ChlaIL in the years when the cover of submerged
vegetation was lowest, whereas a shift occurred to a clearwater state in the
27. Clear and Turbid Water in Shallow Norwegian Lakes
367
following years with Chla concentrations less than 20 I!g ChlaiL. The average
Secchi depth during the growth season increased from less than 0.6 m in the turbid
phase to 1.2-2.5 m in the clearwater phase (Blindow et al., 1993). Turbidity
simultaneously decreased from greater than 22 to less than 8 JTV units. During
these years, total P spontaneously changed from 50-70 I!g PIL before the reinvasion of submerged vegetation to less than 35 I!g PIL aftelward. This gives a
Chlaltotal P ratio of 0.49 and 0.46 during the turbid state and 0.35, 0.43, 0.46, and
0.36 in the clear state. This indicates that the Chlaltotal P ratio is less useful before
the phytoplankton biomass and nutrient concentrations have stabilized in a new
state.
It may be argued that water with 20-30 I!g ChlaIL can be perceived as turbid
when assessing lakes with total P concentrations as low as to 20-30 I!g PIL. In
fact, these are the maximum Chla levels expected in such lakes (Faafeng and
Hessen, 1993). As evidenced by Figures 27.1 and 27.2, clear lakes may contain
Chla concentrations between one-third and one-tenth of the maximum concentrations (e.g., 2-10 I!g ChlalL), at this P level.
More investigations in shallow lakes are needed to obtain a better statistical
relationship between total P, chlorophyll, and transparency, especially in mesoeutrophic lakes with low vegetation cover. Turbidity caused by factors other than
phytoplankton biomass should also be taken into consideration.
Acknowledgments. The authors thank Hanne Edvardsen, Bod0, and a number of
colleagues at NIVA for their assistance and support during macrophyte registration
and sampling and analysis of water samples. Identification and enumeration of
phytoplankton were carried out by P. Brettum, and the zooplankton was analyzed
by D.O. Hessen and J.E. L0vik. Sincere thanks are also due to Ingrid Blindow for
valuable comments. The National Eutrophication Survey of Norwegian Lakes is
financed by the Norwegian State Pollution Authority.
References
Balls, H.; Moss, B.; Irvine, K. The effects of high nutrient loading on interactions between
aquatic plants and phytoplankton. Verh. Int. Verein. Limnol. 22:2912-2915; 1985.
Blindow, 1.; Andersson, G.; Hargeby, A; Johansson, S. Long-term pattern of alternative
stable states in two shallow eutrophic lakes. Freshwat. BioI. 30:159-167; 1993.
Canfield, D.E., Jr.; Langeland, K.A.; Linda, S.B.; Haller, w.T. Relations between water
transparency and maximal depth of macrophyte colonization in lakes. J. Aquat. Plant
Manage. 23:25-28; 1985.
Chambers, P.A; Kalff, J. Depth distribution and biomass of submersed aquatic macrophyte
communities in relation to Secchi depth. Can. J. Fish. Aquat. Sci. 42:701-709; 1985.
Faafeng, B.A.; Hessen, D.O. Nitrogen and phosphorus concentrations and N:P ratios in
Norwegian lakes: perspectives on nutrient limitations. Verh. Int. Verein. Limnol. 25(1):
465-469; 1993.
Grimm, M.P. Northern pike (Esox lucius L.) and aquatic vegetation, tools in the management of fisheries and water quality in shallow waters. Hydrobiol. Bull. 23:61-67; 1989.
Hargeby, A; Andersson, G.; Blindow, 1.; Johansson, S. Trophic web structure in a shallow
lake during dominance shift from phytoplankton to submerged macrophytes. Hydrobiologia 279/280:83-90; 1994.
368
B.A Faafeng and M. Mjelde
Irvine, K.; Balls, H.; Moss, B. The enterostracan and rotifer communities associated with
submerged plants in the Norfolk Broadland-effects of plants and species composition.
Int. Rev. Ges. Hydrobiol. 75:121-141; 1990.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; S\ilndergaard, M.; Mortensen, E.; Sortkjrer, 0.;
Olrik, K. Fish manipulation as a lake restoration tool in shallow, eutrophic, temperate
lakes. 2: Threshold levels, long-term stability and conclusions. Hydrobiologia 200/201:
219-227; 1990.
Mazumder, A Phosphorus-chlorophyll relationships under contrasting herbivory and thermal stratification: predictions and patterns. Can. J. Fish. Aquat. Sci. 51 :390-400; 1994.
Mjelde, M.; Faafeng, B. Ceratophyllum demersum (L.) hampers phytoplankton development in some small Norwegian lakes over a wide range of phosphorus level and
geographic latitude. Freshwat. BioI. 37:355-365; 1997.
Phillips, G.L.; Eminson, D.E; Moss, B. A mechanism to account for macrophyte decline in
progressively eutrophicated freshwaters. Aquat. Bot. 4: 103-126; 1978.
Riley, E.T.; Prepas, E.E. Comparison of the phosphorus-chlorophyll relationships in mixed
and stratified lakes. Can. J. Fish. Aquat. Sci. 42:831-835; 1985.
Sand-Jensen, K.; S\ilndergaard, M. Phytoplankton and epiphyte development and their
shading effect on submerged macrophytes in lakes of different nutrient status. Int. Rev.
Ges. Hydrobiol. 66:529-552; 1981.
Scheffer, M. Alternative stable states in eutrophic shallow fresh water systems: a minimal
model. Hydrobiol. Bull. 23:73-84; 1989.
Scheffer, M. Multiplicity of stable states in freshwater systems. Hydrobiologia 200/201:
474-486; 1990.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternate equilibria in
shallow lakes. Trends Ecol. Evol. 8 :275-279; 1993.
Scheffer, M.; Van den Berg, M.; Breukelaar, A; Breukers, c.; Coops, H.; Doef, R; Meijer,
A.-L. Vegetated areas in turbid shallow lakes. Aquat. Bot. 49: 193-196; 1994.
Schriver, P.; B\ilgestrand, J.; Jeppesen, E.; S\ilndergaard, M. Impact of submerged macrophytes on fish-zooplankton-phytoplankton interactions: large-scale enclosure experiments in a shallow eutrophic lake. Freshwat. BioI. 33:255-270; 1995.
Timms, RM.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a freshwater
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
van Donk, E.; Gulati, RD. Transition of a lake to turbid state six years after biomanipulation: mechanisms and pathways. Wat. Sci. Techn. 32: 197-206; 1995.
van Donk, E.; Grimm, M.P.; Gulati, RD.; Heuts, G.M.; de Kloet, W.A; van Liere, L. First
attempt to apply whole lake food-web manipulation on a large scale in the Netherlands.
Hydrobiologia 200/201 :291-301; 1990.
van Donk, E.; Gulati, R.D.; Iedema, A.; Meulemans, J.T. Macrophyte-related shifts in the
nitrogen and phosphorus content in the different trophic levels in a biomanipulated
shallow lake. Hydrobiologia 251 :19-26; 1993.
Vant, W.N.; Davies-Colley, R.I.; Clayton, J.S.; Coffey, B.T. Macrophyte depth limits in
North Island (New Zealand) lakes of different clarity. Hydrobiologia 137:55-60; 1986.
Wium-Andersen, S. Allelopathy among aquatic plants. Erg. Limnol. 27: 167-172; 1987.
28.
Macrophytes and Thrbidity in Brackish Lakes
with Special Emphasis on the Role of
Top-Down Control
Erik Jeppesen, Martin S0ndergaard, Jens Pedler Jensen,
Eva Kanstrup, and Birgitte Petersen
Introduction
Evidence from both empirical studies (Canfield et aI., 1984; Jeppesen et aI.,
1990; Faafeng and Mjelde, this volume, Chapter 27) and numerous experimental field studies (see e.g., Gulati et aI., 1990; van Donk et aI., 1990; Mortensen
et aI., 1994) indicates that in freshwater lakes extensive growth of submerged
macrophytes may lead to clearwater conditions, even at high nutrient concentrations. Several factors seem to be involved, including both increased
zooplankton grazer control and nutrient constraint on phytoplankton, alterations in the physical environment that result in less wind-induced and fishinduced resuspension, and possibly also allelophatic effects (Jeppesen et aI.,
1990; Moss, 1990; Scheffer et aI., 1993). A cross-analysis of survey data from
35 Danish brackish lakes revealed a significant decrease in Secchi depth with
increasing concentrations of total phosphorus (TP); in contrast to freshwater
lakes, however, transparency was independent of whether submerged macrophytes were present at high density (Jeppesen et aI., 1994). Similarly, Moss
(1994) found that nutrient-rich brackish lakes with extensive growth of submerged macrophytes tend to be in a turbid state. By using both empirical data
and field experiments conducted in several brackish and freshwater shallow
Danish lakes, we examine here how differences in top··down control may
influence the turbidity of freshwater and brackish lakes in the macrophyte
state.
369
370
E. Jeppesen et al.
Materials and Methods
The analysis is based on survey data from 50-100 freshwater lakes and 35
brackish lakes. Fish population estimates are based on fish caught overnight in gill
nets (3 x 1.5-m sections, 14 different mesh sizes from 6.25 to 75 mm) expressed as
catch per unit effort (CPUE = fish/netl19 h). Most of the sampling procedures and
methods are described by Jeppesen et al. (1994) and Aaser et ai. (1995), and only
additional methods are presented here. Leptodora kindti was counted on zooplankton samples taken at equidistant intervals from the surface to the bottom at
one-three stations in the pelagic zone. Chaoborus spp. density was estimated from
their abundance (nlm 2 ) in sediment samples collected during the day in autumn or
spring, and we assumed that they were evenly distributed in the pelagic zone at
night. Between 3-10 samples were taken with a Kajak sampler (diameter, 5.2 cm)
in each lake on one-five occasions during winter or spring. The estimate is thus
conservative, as summer densities of Chaoborus are higher (e.g., Christoffersen,
1990).
Sampling of Neomysis integer in shallow Lake 0rslevkloster (40 ha; mean
depth, about 2 m; max depth, about 4 m; salinity, 2-4%0) was conducted at 16
stations by vertical hauling with a 0.5-mm net (diameter, 0.5 m) according to
Aaser et al. (1995). Fish sampling in this lake was conducted by using 1.5 x 32-m
sinking gill nets (eight 4-m sections; each including I-m sections of 6.25-, 8-, 10-,
and 12.5-mm mesh size, respectively), four nets being placed overnight in the
littoral zone running parallel to the shore and two in the pelagic at a mid-lake
station. Physico-chemical data were obtained by sampling at mid-lake stations
(pooled samples from the entire water column).
Results and Discussion
A plot of TP versus Secchi depth in Danish lakes shows that nutrient -rich brackish
lakes are turbid even when macrophyte densities are high (Fig. 28.1). Chlorophyll a was significantly linearly related to TP and unrelated to submerged macrophyte coverage (Fig. 28.2). These results indicate that zooplankton grazing on
phytoplankton is unaffected by the presence or absence of macrophytes in brackish lakes. In eutrophic freshwater lakes, by contrast, the macrophyte state is generally associated with high transparency (Fig. 28.1) and usually, but not always (Meijer et
al., 1994), with a high zooplankton/phytoplankton biomass ratio and hence a potentially high grazing pressure on phytoplankton (Moss et aI., 1994; Jeppesen et aI.,
1997; Van den Berg et aI., this volume, Chapter 25). Lower zooplankton grazing in
brackish lakes may partly reflect differences in zooplankton community structure.
Although large-bodied Daphnia (e.g., D. magna), which playa key role in grazer
control of phytoplankton in freshwater lakes (Carpenter and Kitchell, 1993),
may become dominant in slightly brackish lakes (Jurgens and Stolpe, 1995), they
most frequently disappear above salinities of 2-4%0 (Jeppesen et aI., 1994; Moss,
1994). Instead, the lakes are dominated by the copepods, Eurytemora spp. and
371
28. Top-Down Control in Brackish Lakes
5~--------------------------------'--~
Freshwater lakes
4
0
~
3 8
3
2 :
\q,
o
~:~... ~:;:.:~:.~...:;:............................................................
0
o
0.1
0.2
0.3
0.4
0.5
0.6
0.7 O.S
0.9
1.0
.".
1.1
0
1.5
Total phosphorus (m9 P 1. 1 )
Figure 28.1. Secchi depth in relation to lake water total phosphorus in shallow freshwater
(upper panel) and brackish lakes (lower panel). 0 lakes with more than 30% submerged
macrophyte coverage; e lakes with a low « 30%) or unknown submerged macrophyte
coverage. Each point represents one lake and is a time-weighted average of all data
collected between May 1 and October I. The broken line indicate,. an exponential curve
developed by Kristensen et al. (1991) on the basis of data from freshwater and brackish
lakes with low submerged macrophyte coverage. (From Jeppesen et aI., 1994. Published
with permission from Kluwer Academic Publishers.)
Acartia spp., and occasionally by rotifers (Jeppesen et aI., 1994), which are
probably less efficient in controlling phytoplankton than large-bodied cladocerans. In addition, the zooplankton/phytoplankton ratio is low in brackish lakes
(Jeppesen et aI., 1994).
Lake 0rslevkloster is an example of how a salinity-mediated shift in trophic
structure may reduce grazer control on phytoplankton in macrophyte-rich brackish
lakes. The lake shifted from a brackish state 0-3%0) dominated by Eurytemora
affinis and rotifers to a slightly brackish state (0.5-1%0) dominated by Daphnia
galeata (Fig. 28.3). Chlorophyll a was 2.5-3.5-fold higher illl the brackish state
1500.,--;:-;----;--;----------,,----,
Macrophyte coverage
-0·30%
o > 300/0
Ol
2:1000
111
'5.
.s:::
Figure 28.2. Summer mean chlorophyll a
versus total phosphorus in some Danish shallow brackish lakes with submerged macrophyte coverage in the range ~30% (e) or
greater than 30% (0).
e
o
500
-,It-••
:2
o
o I..~-
o
500
1000
1500
Total phosphorus (lJg P 1. 1 )
2000
372
E. Jeppesen et al.
(1993 and 1994) during summer than in the slightly brackish state and 17-23-fold
higher during autumn. Correspondingly, Secchi depth was 20-50% and 10--12%
lower during summer and autumn, respectively (Fig. 28.4). Despite the fact that
external loading did not change (Viborg County, 1995), lake water TP was highest
in the brackish state, which is probably due to internal loading caused by FeS
formation and a resultant release of iron-bound phosphorus as demonstrated in the
nearby Hjarbrek Fjord (H. Jensen, unpublished results). The higher chlorophyll a in
the brackish state may therefore partly reflect the higher P concentration. However, the
chlorophyll a!TP ratio during summer and autumn also tended to be higher, which
may indicate lower zooplankton grazing on phytoplankton. This is supported by the
substantially lower zooplankton/phytoplankton biomass ratio in the brackish state
during summer (0.02-0.06 versus 0.49 in the slightly brackish state) (Fig. 28.3). The
shifts in turbidity and zooplankton/phytoplankton biomass ratios in Lake 0rslevkloster thus follow the predictions of the established empirical relations, but we
cannot, however, exclude the possibility that the shift is caused by other factors.
Changes in the recruitment of fish unrelated to changes in salinity may, for instance,
also have played a role, but sufficient data to elucidate this are not available.
The low zooplankton/phytoplankton biomass ratio in the brackish state in Lake
0rslevkloster (Fig. 28.3) and in other Danish brackish lakes (Jeppesen et al., 1994)
cannot simply be explained by lack of edible phytoplankton. Green algae and
diatoms, which are a common food source for the dominant zooplankton (Eurytemora affinis and rotifers), are abundant in most eutrophic brackish lakes (Balls
et aI., 1993; J.P. Jensen et al., unpublished observation), as well as in Lake
0rslevkloster (Nielsen, 1995). Another explanation is top-down control via invertebrates and fish. In North European eutrophic brackish lakes, Neomysis integer is
a major invertebrate predator (Irvine et aI., 1990; Moss, 1994; Aaser et al., 1995),
and mysid density increases with increasing TP, particularly above 400 /lg PIL
(Fig. 28.5), reaching densities as high as 131L (Jeppesen et aI., 1994). The marked
increase in mysid density coincides with a shift in the fish community from
dominance by roach (Rutilus rutilus), perch (Perea jluviatilis), whitefish (Coregonus spp.), smelt (Osmerus spp.), etc., to exclusive dominance by small-sized
sticklebacks (Gasterosteus spp.) (Jeppesen et aI., 1994). The latter coexist with N.
integer, probably because unlike the larger fish, sticklebacks prey selectively on
smaller stages of mysids rather than on ovigorous females (Jeppesen et aI., 1994;
Kanstrup, 1996). By contrast, the maximum density of the pelagic invertebrate
predators in freshwater lakes (Leptodora kindti and Chaoborus spp.) occurs at
100--200 /lg P/L, and they almost disappear at high TP (Fig. 28.5). This is probably
due to increased fish predation because large-sized fish such as roach and bream
(Abramis brama) dominate in north European hypertrophic lakes (Persson et aI.,
1988; Jeppesen et al., 1990). Although N. integer is omnivorous (Arndt and
Jansen, 1986) as opposed to the more strict carnivorous Chaoborus and Leptodora
found in freshwater lakes, the predation pressure on zooplankton by pelagic
invertebrate predators is probably higher in hypertrophic brackish lakes than in
comparable freshwater lakes because of the very high predator density that
Neomysis may reach in the brackish lakes.
28. Top-Down Control in Brackish Lakes
i
%%
1993
1994
1993
1995
1994
373
~f~~Vr~~::::~:n:dS
Daphnia sp.
Calanoid
copepods
1995
Figure 28.3. Zooplankton biomass and zooplankton/phytoplankton biomass ratio (A) and
percentage of biomass accounted for by the various zooplankton groups (B) in shallow
Lake 0rslevkloster in 1993, 1994, and 1995. No quantitative data are available for 1986,
but high density of Daphnia hyalina was observed in littoral fauna samples (Viborg County,
1988), thus indicating that the lake was then in a cladoceran state.
1986
1993
1994
1995
350.------------------------.
~
300
3 250
t1l
200
o
""~
0...
I-
ili
} 0.6
a.
eo
:c
u
1986
1993
1994
1995
1986
1993
1994
1995
Figure 28.4. Spring (Jan I-May 1), summer (May l-Oct I) and autumn (Oct I-Jan I)
mean (±SE) values of chlorophyll a, Secchi depth, total phosphorus and chlorophyll a/total
phosphorus ratio in Lake 0rslevkloster during 4 years differing in salinity (upper panel).
374
E. Jeppesen et al.
Pelagic invertebrate predators
A
Plankli-benlhivorous fish
c==---------~~-----------C~
Freshwater 0
Brackish water IlilI
Chaoborus •
Leptodora 0
Neomysisliil
o
1
234
Abundance (no. 1. 1)
o
2
4
6
8
10
CPUE (kg net 1 )
0
100
200
300
400
CPUE (no. 1. 1)
Figure 28.S. Summer (May I-Oct I) mean abundance of some invertebrate predators in
Danish freshwater and brackish lakes (A). CPUE of planktivorous fish in terms of biomass
(B) and number (C) (multiple mesh size gill netting in late JUly-August, 14 different mesh
sizes, 6.25-75 mm), each versus summer mean lake water total phosphorus.
Planktivorous fish significantly affect zooplankton abundance, composition,
and the zooplankton/phytoplankton biomass ratio in Danish shallow eutrophic
freshwater lakes (Jeppesen et aI., 1997). The biomass of planktivorous fish caught
in multiple mesh-sized gill nets in brackish lakes is lower than in freshwater
lakes, particularly at higher TP, when small-sized sticklebacks dominate (Fig.
28.5; for discussion about net selectivity, see Jeppesen et aI., 1994). In terms of
numbers, however, CPUE was not lower at high TP concentrations. The data from
the freshwater lakes therefore suggest that the predation pressure on zooplankton
should be high also in eutrophic brackish lakes (Jeppesen et aI., 1994). In addition,
sticklebacks produce offspring several times during the summer and autumn.
Predation pressure on zooplankton by fish fry is particularly high (e.g., He and
Wright, 1992; S0ndergaard et aI., 1997; Jeppesen et aI., 1997), suggesting that
there is more likely a continuously high fish predation pressure on zooplankton in
eutrophic brackish lakes than in comparable freshwater lakes. This idea of
higher invertebrate and fish predation is further supported by the lower zooplankton/phytoplankton biomass ratio in brackish lakes (Fig. 28.3; Jeppesen
et aI., 1994).
Aggregation of G. aculeatus and N. integer in the littoral zone (Arndt and
Jensen, 1986) may be a contributory factor to the higher turbidity of macrophyterich brackish lakes, as it may diminish the ability of the pelagic zooplankton to use
macrophytes as a daytime refuge. In Lake 0rslevkloster, the gill net catch of
sticklebacks was about sevenfold higher in the littoral zone than in the open water
during November-August, and 1O-25-fold higher during the summer (Kanstrup,
1996; Fig. 28.6). After August, the pattern changed, however, with the number of
stickleback caught being highest in the open water. Correspondingly, the annual
mean concentration of N. integer was 120-fold higher in the littoral zone than in
the pelagic zone (Fig. 28.6), or 30-fold higher per unit area. An experiment
involving partial harvesting of macrophytes in the littoral zone of the same lake
28. Top-Down Control in Brackish Lakes
14
12
ci
.s
Neomysis integer
_
_
Littoral
Pelagial
300
A
10
:;:.
8
ci
375
Gasterosteus acu/eatus
B
250
~ 200
.s 150
6
w
~ 100
4
()
50
2
1992
1993
Figure 28.6. Seasonal variation in the density (±SE) of Neomysis integer (A) and gill net
CPUE of three-spined stickleback (Gasterosteus aculeatus) (B) in the littoral and pelagic
zone of Lake 0rslevkloster in 1992/93.
showed an 80% higher mysid density inside the plant beds than at similar depths
outside (Petersen, 1994). The suggested low refuge effect of macrophytes in
brackish lakes may contribute to the low zooplankton control of phytoplankton
and hence to the high turbidity of macrophyte-rich lakes, as the possibility of
seeking daytime refuge in the vegetation has been shown to be a key factor for the
survival of pelagic cladocerans in macrophyte-rich freshwater lakes with a high
density ofplanktivorous fish (Jeppesen et aI., this volume, Chapter 5; Lauridsen et
al., this volume, Chapter 13).
Nutrient release by N. integer may also contribute to a different response of the
two lake types. Experiments conducted in two shallow brackish lakes thus showed
a considerably higher TP in enclosures containing mysids than in controls devoid
of mysids (Aaser et aI., 1995; Nielsen, 1995). The results indicated that mysids
enhance nutrient release from the sediment, perhaps because some of the nutrients
ingested when feeding on sediment detritus and benthic invertebrates are subsequently released to the pelagic. This, in tum, may stimulate phytoplankton
growth, thereby contributing to the low Sec chi depth in brackish lakes.
Although we are beginning to understand the mechanisms behind the high
turbidity of eutrophic macrophyte-rich brackish lakes, more research is needed
before any firm conclusions can be drawn. Further studies are important not only
from a basic science point of view but also with regard to lake management.
Thus, the difference in trophic structure and dynamics of the two different lake
types has important implications when transferring the ecotechnological restoration methods known from freshwater lakes to brackish lakes (Jeppesen et
al., 1994; Moss, 1994).
Acknowledgments. We thank the Danish Counties for providing access to some of
the data used in the analyses. The assistance of the technical staff of the National
Environmental Research Institute, Silkeborg, is gratefully acknowledged. We also
376
E. Jeppesen et aI.
thank Mark Hoyer and Marten Scheffer for valuable comments. The study was
supported by the Centre for Freshwater Environmental Research.
References
Aaser, H.F.; Jeppesen, E.; Sf,1ndergaard, M. Seasonal dynamics of the mysid Neomysis
integer and its predation on the copepod Eurytemora affinis in a shallow hypertrophic
brackish lake. Mar. EcoI. Prog. Ser. 127:47-56; 1995.
Arndt, E.A.; Jansen, W. Neomysis integer (Leach) in the Chain of Boddens south of
Darss/Zingst (Western Baltic). Ecophysiology and population dynamics. Ophelia 4:115; 1986.
Balls, M.; Moss, B.; Phillips, G.L.; Irvine, K.; Stansfield, H. The changing ecosystem of a
shallow, brackish lake, Hickling Broad, Norfolk II. Long-term trends in water chemistry
and ecology and their implications for restoration of the lake. Freshwat. BioI. 29: 141165; 1993.
Canfield, D.E.; Shireman, J.V.; Colle, D.E.; Haller, w.T.; Watkins, C.E.; Maceina, M.J.
Prediction of chlorophyll a concentrations in Florida lakes: importance of aquatic
macrophytes. Can. J. Fish. Aquat. Sci. 41 :497-501; 1984.
Carpenter, S.R.; Kitchell, J.F., eds. The trophic cascade in lakes. New York: Cambridge
University Press; 1993.
Christoffersen, K. Evaluation of Chaoborus predation on natural populations of herbivorous zooplankton in a eutrophic lake. Hydrobiologia 200/201:459-466; 1990.
Gulati, R.D.; Lammens, E.H.R.R.; Meijer, M.-L.; van Donk, E. Biomanipulation, tool for
water management. Hydrobiologia 2001201: 1-628; 1990.
He, X.; Wright, R. An experimental study of piscivore-planktivore interactions: population
and community responses to predation. Can. J. Fish. Aquat. Sci. 49: 1176-1185; 1992.
Irvine, K.; Bales, M.T.; Moss, B.; Stansfield, J.H.; Snook, D. Trophic relations in Hickling
Broad-a shallow and brackish eutrophic lake. Verh. Int. Verein. Limnol. 24:576-579;
1990.
Jeppesen, E.; Jensen, J.P.; Kristensen, P.; Sf,1ndergaard, M.; Mortensen, E.; Sortkjrer, 0.;
Olrik, K. Fish manipUlation as a lake restoration tool in shallow, eutrophic, temperate
lakes 2: Threshold levels, long-term stability and conclusions. Hydrobiologia 200/201:
219-227; 1990.
Jeppesen, E.; Sf,1ndergaard, M.; Kanstrup, E.; Petersen, B.; Eriksen, R.B.; Hammershf,1j, M.;
Mortensen, E.; Jensen, J.P.; Have, A. Does the impact of nutrients on the biological
structure and function of brackish and freshwater lakes differ? Hydrobiologia 275/276:
15-30; 1994.
Jeppesen, E.; Jensen, J.P.; Sf,1ndergaard, M.; Lauridsen, T.L.; Junge Pedersen, L.; Jensen, L.
Top-down control in freshwater lakes: the role of nutrient state, submerged macrophytes
and water depth. Hydrobiologia 3421343:151-164; 1997.
Jiirgens, K.; Stolpe, G. Seasonal dynamics of crustacean zooplankton, heterotrophic flagellates and bacteria in a shallow eutrophic lake. Freshwat. BioI. 33:27-38; 1995.
Kanstrup, E. Trepigget hundestejles Gasterosteus aculeatus L. betydning for de biologiske
interaktioner i en lavvandet, eutrof brakvandssf,1 (in Danish). [The influence of threespined stickleback Gasterosteus aculeatus on the biological interactions in a shallow
eutrophic brackish lake.] MSc thesis, National Environmental Research Institute and the
University of Aarhus, Aarhus; 1996.
Kristensen, P.; Jensen, J.P.; Jeppesen, E. Simple empirical lake models. In: Nitrogen and
phosphorus in fresh and marine water. Danish Environmental Protection Agency Abstracts,
Copenhagen; 125-145; 1991.
Meijer, M-L.; Jeppesen, E.; van Dank, E.; Moss, B.; Scheffer, M.; Lammens, E.; van Nes,
E.; van Berkum, J.A.; de Jong, G.J.; Faafeng, B.A.; Jensen, J.P. Long-term response to
fish-stock reduction in small shallow lakes: interpretation of five-year results of four
28. Top-Down Control in Brackish Lakes
377
biomanipulation cases in the Netherlands and Denmark. Hydrobiologia 275/276:457466; 1994.
Mortensen, E.; Jeppesen, E.; S!/indergaard, M.; Kamp Nielsen, L., eds. Nutrient dynamics
and biological structure in shallow freshwater and brackish lakes. Hydrobiologia 2751
276:1-507; 1994.
Moss, B. Engineering and biological approaches to the restoration from eutrophication of
shallow lakes in which aquatic plant communities are important components. Hydrobiologia 2001201:367-378; 1990.
Moss, B. Brackish and freshwater lakes--different systems or variations on the same
theme? Hydrobiologia 2751276:367-378; 1994.
Moss, B.; McGowan, S.; Carvalho, L. Determination of phytoplankton crops by top-down
and bottom-up mechanisms in a group of English lakes, the West Midland meres.
Limnol. Oceanogr. 39:1020-1029; 1994.
Nielsen, F. Gra:sning af copepoder Eurytemora affinis i to eutrofe S!/ier (in Danish). [Eu1)'temora affinis grazing in two eutrophic lakes.] MSc thesis, National Environmental
Research Institute, Silkeborg, and the University of Aarhus, Aarhus; 1995.
Persson, L.; Anderson, G.; Harnrin, S.F.; Johansson, L. Predation regulation and primary
production along the productivity gradient of temperate lake ecosystems. In: Carpenter,
S.R., ed. Complex interactions in lake communities. New York: Springer-Verlag; 1988:
45-65.
Petersen, B. Neomysis integers !/ikologiske rolle i en lavvandet, eutrof brakvandss!/i (in
Danish). [The ecological role of Neomysis integer in a shallow eutrophic brackish lake.]
MSc thesis, National Environmental Research Institute, Silkeborg, and the University of
Aarhus, Aarhus; 1994.
Scheffer, M.; Hosper, S.H.; Meijer, M.-L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends Ecol. Evol. 8:275-279; 1993.
S!/indergaard, M.; Jeppesen, E.; Berg, S. Pike (Esox lucius L.) stocking as a biomanipulation
tool. 2. Effects on lower trophic levels in Lake Lyng (Denmark). Hydrobiologia 3421
343:319-325; 1997.
van Donk, E.; Grimm, M.P.; Gulati, R.D.; Klein, J.P.G. Whole-lake food-web manipulation
as a means to study community interactions in a small ecosystem. Hydrobiologia
200/201 :275-289; 1990.
Viborg County. Milj!/itilstand i 0rslevkloster S!/i 1986-1987 (in Danish). [The environmental state of Lake 0rslevkloster 1986-1987.] Viborg, Denmark; 1988.
Viborg County. 0rslevkloster S!/i 1994. Belastning, fysisk-kemiske forhold, vegetation samt
fiskebestand (in Danish). [Lake 0rslevkloster 1994. Loading, physico-chemical interactions, vegetation and fish stock.] Viborg, Denmark; 1995.
3. Interdisciplinary Discussions
29.
Structuring Role of Macrophytes in Lakes:
Changing Influence Along Lake Size and
Depth Gradients
Avital Gasith and Mark V. Hoyer
Introduction
Emergent, floating-leaved, and submergent macrophytes grow in the littoral region of most lakes. These aquatic macrophytes are influenced by geomorphology,
environmental conditions, and biotic interactions (Sculthorpe, 1967; Hutchinson,
1975), while exerting their own influence on the lake environment and biota
(Carpenter and Lodge, 1986; Engel, 1988). The capacity of macrophytes to provide a substrate for colonization of algae and invertebrates (Sozska, 1975; Cattaneo and Kalff, 1980; Dvorak and Best, 1982; Cattaneo, 1983; Morin, 1986;
Schram et aI., 1987; Miller et al., 1989), to affect water and sediment chemistry as
well as other lirnnological conditions (Carpenter and Gasith, 1978; Prentki et aI.,
1979; Jaynes and Carpenter, 1986), and to influence biogeochemical cycles and
productivity (Wetzel and Hough, 1973; Godshalk and Wetzel, 1978; Wetzel, 1979;
Carpenter, 1980; Cattaneo and Kalff, 1980; Carpenter, 1983; Wetzel, 1990) and
biotic interactions (Crowder and Cooper, 1982; Heck and Crowder, 1991; Schriver
et al" 1995; see also this volume) is well recognized. The understanding of the role
of macrophytes in lacustrine systems is based mostly on process studies, smallscale investigations (ponds, test plots), observations in small lakes, and modeling
(Carpenter and Lodge, 1986). It is intuitively obvious that the influence of macrophytes in most small or shallow aquatic systems is proportional to their abundance
(density, biomass, or extent of cover) and productivity. Little is known about the
role of macrophytes in situations in which they are less conspicuous, as in large
381
382
A. Gasith and M.V. Hoyer
deep lakes. Danehy et al. (1991), Gasith (1991), and Gasith and Gafny (this
volume, Chapter 24) argue that the potential influence oflittoral resources, including those provided by macrophytes to the biotic functioning of large deep lakes,
has been overlooked. Both abundance and productivity of macrophytes vary about
two orders of magnitude among lakes of different trophic levels (Carpenter, 1983),
regardless of lake size. It is less clear, however, how the role of macrophytes varies
in lakes of similar trophic status that differ in size and depth. The purpose of this
discussion is to assess how the potential structuring role of macrophytes can
change along lake size (surface area) and depth gradients.
We first point out the inherent difficulty in the terminology used to describe a
lake size; we then consider the factors that interact with lake size and depth and
affect plant growth; and finally, we assess the changing role of macrophytes along
lake size and depth gradients.
The macrophyte-epiphyte complex is functionally inseparable. Whenever we
generally use the term macrophytes, it is inclusive of their epiflora. For sake of the
required brevity, we also fail to distinguish among the different macrophyte types
and growth forms, despite evidence for possible type or growth form-specific
effects as well as effects of mixed plant associations (Emery, 1978; Guillory et ai.,
1979; Eadie and Keast, 1984; Conrow et al., 1990; Dionne and Folt, 1991; Lillie
and Budd, 1992; Chick and Mclvor, 1994).
Large Versus Small and Deep Versus Shallow
Lakes are commonly categorized as small or large and shallow or deep despite
lack of clear-cut morphological definitions. Generally, large lakes tend to be
deeper and have longer retention times than small lakes. Only large deep lakes
have truly pelagic communities that are usually more important in the overall
cycling and production processes than the littoral zone and bottom communities
(Tilzer, 1990). The term shallow is often associated with lakes that do not thermally
stratify and where continuous sediment-water interaction makes internal nutrients
cycling more efficient than in deeper lakes that stratify. This definition ignores the
important presence of aquatic macrophytes. Thus, a definition more pertinent to the
aim of this discussion is that shallow lakes are those whose bottom is significantly
covered by submerged macrophytes (Moss, 1995). In general, large deep lakes Ilave
less aquatic macrophytes than small shallow lakes, with the exception that highly
turbid, shallow lakes may be devoid of submerged vegetation.
Factors Affecting Plant Growth: Interaction with Lake Size
and Depth
Here, we consider lake size (surface area) and depth on a relative scale in connection to the potential growth of aquatic macrophytes. Unless stated otherwise, we
assume similar growth conditions for the lakes compared, except for those arising
from the gradients in surface area and depth.
29. The Importance of Lake Size and Depth
383
Comparison of the role of macrophytes along lake size and depth gradients is
complicated because plant development is variable even in lakes of similar morphometery (Sculthorpe, 1967; Hutchinson, 1975). Several site-specific environmental factors affecting the abundance and distribution of aquatic macrophytes in
lakes have been identified. These include climatic factors such as irradiance,
temperature, wave action generated by winds, size and edaphic features of the
catchment basin that affect nutrient loading and general water chemistry, and
biotic factors of grazing by invertebrates, fish, and birds. We limit our discussion
to those factors that interact with lake size and depth.
Light availability and wave action are directly and indirectly influenced by
morphometric features. Due to exponential light attenuation in water, depth is one
of the most critical environmental factors determining the lakeward growth of
macrophytes and their species richness (Hutchinson, 1975; Chambers and Kalff,
1985; Duarte et aI., 1986). As a general rule of thumb, submerged macrophytes
will grow to a depth of two to three times the Secchi depth (Canfield et aI., 1985;
Chambers and Kalff, 1985). Thus, macrophyte growth will be limited in lakes with
small or large surface areas where the majority of lake bottom exceeds the above
Secchi depth. Additionally, even if a lake is physically shallow and does not
thermally stratify (i.e., 1-2 m, mean depth), if the Secchi depth is less than 0.5 m
there is a strong probability that submerged aquatic macrophytes will be absent.
With some exceptions, a depth range between 10 and 15 m appears to be a limit for
most angiosperms. Lakes in which most of the basin is deeper than 10-15 mare
not expected to have abundant submerged aquatic macrophytes. Emergent and
floating-leaved aquatic macrophytes seldom grow in waters exceeding a depth of
3 m (Canfield and Hoyer, 1992). Climatic differences associated with lake latitude
appear to have a strong influence on the relationship between depth distribution of
submerged plants and water transparency (Duarte and Kalff, 1987). At low
latitudes, angiosperms colonize deeper and reach maximum biomass at greater
depth than those growing in lakes of similar transparency at higher latitudes.
Warmer water, greater irradiance, and longer growing period in lower latitude
lakes may account for the difference.
Basin slope (square root of the area divided by mean depth; Hakanson, 1981),
surface area, and basin configuration are among the most important morphological
features that influence the potential development of macrophytes in lakes (Pearsall, 1917; Spence, 1982; Duarte and Kalff, 1986). These factors interact directly
and indirectly with other environmental factors such as light, nutrient availability,
substrate characteristics, and wind-generated erosion to detennine the site-specific
extent of plant development and macrophyte types.
Maximum biomass of submerged macrophytes is inversely related to slope
(Duarte and Kalff, 1986). The probable reasons for this relation is the difference in
the relative area suitable for plant growth and in sediment stability and quality
between gently and steeply sloped littoral zones. The area oflittoral zone available
for emergent growth declines with increasing slope of the basin. In addition,
steep-sided basins are areas of erosion and sediment transport (Pearsall, 1917;
Hakanson, 1977), whereas nearshore regions of gently sloped basins are sites of
384
A. Gasith and M.V Hoyer
accretion of fine, relatively more stable, and nutrient-richer sediment, where
macrophytes can become established. Pearsall (1920) demonstrated that the variation in the quantity and quality of silts largely controls the distribution of submerged vegetation. Thus, irrespective of lake size, steep-sided lakes will have
lower cover and biomass of submerged macrophytes than lakes with gently sloped
basins.
A large lake has a long fetch and a greater wave energy than a smaller lake.
Exposure to waves can directly and indirectly affect plant distribution and abundance in lakes (Keddy, 1983; Chambers, 1987; Coops et al., 1991). Wave action
and currents also affect sediment transport and distribution in lakes (DavidsonArnottand Pollard, 1980; Keddy, 1982), concomitantly affecting the distribution of aquatic plants (Spence, 1982). Unless physically protected, points and
shallows where wave energy is highest tend to be swept clean of fine sediments
(Lorang and Stanford, 1993) and have little or no growth of macrophytes. Bays
and areas below the wave-mixed depth tend to silt in providing more stable
sediments, suitable for the establishment of macrophytes (Pearsall, 1929).
Waves and strong currents can also retard vegetation growth by exerting a
mechanical stress on the plants (Hutchinson, 1975; Coops et aI., 1991). High
concentration of suspended solids generated by wind mixing of bottom sediments (Kristensen et aI., 1992) can limit light for plant growth particularly in
large, shallow, unstratified lakes, whereas in stratified lakes suspended particles tend to settle out of the mixed layer (Osgood, 1988). High wave energy,
currents, and turbidity in the shallows often restrict macrophyte growth in
large deep lakes to protected bays and coves (Duarte et aI., 1986). Overall,
lakes with large surface areas and longer fetch are expected to have fewer
vegetated littoral regions in relation to the amount of open water than smaller
lakes (Rounsefell, 1946; Spence, 1982).
Lakes with a large surface area tend to be deeper than smaller lakes (a positive
correlation exists between lake area and mean depth; Duarte et aI., 1986). The
cover and biomass of submerged macrophytes are expected to decline with increasing lake size if only for the reason that larger lakes have greater proportion of
area below the compensation depth for macrophytes. In analyzing l39 lakes,
Duarte et al. (1986) indeed found that the percentage surface area covered by
submerged plants is not a constant proportion of the lake area but tends to be
smaller in bigger lakes. Rather surprising, however, was their finding that emergent macrophytes colonized on average a constant proportion (7%) of the lake
area regardless of the size of the lake. A similar relation was reported for Polish
lakes, showing that emergents covered a relatively narrow range of lake surface
areas (9.3-12.3%; Planter, 1973). This contradicts the expectation of declining
growth of macrophytes with increasing fetch and greater wave action in the littoral
zone (Spence, 1982). Duarte et al. (1986) suggested that a greater number of
sheltered bays and floodplains in larger lakes where macrophytes can grow compensates for decreases in vegetation caused by greater wave action. If this is
indeed so, it is apparently sufficient to compensate for the lower growth of
emergents in shoals of large lakes but not of submerged macrophytes. Duarte et al.
29. The Importance of Lake Size and Depth
385
(1986) concluded that on average submerged macrophytes are more important in
small lakes and emergent plants will become more important with increasing lake
size. It should be pointed out, however, that an opposite trend of a transition from
submergents' dominance to that of emergent vegetation is part of the natural
process of lake succession, which is most accelerated in smaIl shallow productive
lakes. The accumulation of refractory macrophyte detritus further limits growth of
submerged macrophytes and hastens the transition to emergent vegetation that is
more tolerant of organic rich sediments (Wetzel, 1979; Carpenter, 1981; Barko and
Smart, 1983).
The proportion of littoral zone areas in a lake declines with increasing depth
and lake size (Gasith, 1991) and increases with increasing shoreline irregularity
(high shore development figure). Therefore, highly irregular lakes may have a
higher proportion of vegetation zones compared with lakes of similar area but with
a more regular shoreline.
The trophic status of lakes is inversely related to mean depth (Vollenweider,
1975; Canfield and Bachmann, 1981). Deep lakes tend to be more oligotrophic
and support lesser growth of aquatic macrophytes than shallow lakes. A study
by Canfield and Hoyer (1992) shows that oligotrophic and mesotrophic lakes
rarely have aquatic macrophyte abundance exceeding 20% volume infested
(PVI), whereas eutrophic and hypereutrophic lakes have the potential to reach
100 PVI. High turbidity may limit growth of submerged macrophytes in these
lakes even though nutrients availability can support extensive growth.
Structuring Role of Macrophytes: Changing Importance Along
Size and Depth Gradients
When established in a lake, aquatic macrophytes can influence the lake ecosystem
in multiple ways (reviewed in Carpenter and Lodge, 1986) and mediate biotic
interactions (Crowder and Cooper, 1982; Savino and Stein, 1982; Diehl, 1988; see
also this volume). The structuring role of macrophytes in a lake ecosystem faIls
into three main categories: (I) limnologicaI effects related to changes in physical
and chemical conditions in the water and sediment; (2) metabolic effects related to
production and processing of organic matter and nutrient cyclmg; and (3) effect on
biotic interactions and community structure related to the role of macrophytes in
providing a structured habitat.
It may be useful to approach the question of how the role of macrophytes
changes along lake size and depth gradients by considering each of the above
categories separately. We suggest that the limnological and metabolic effects
of macrophytes in lakes diminish with increasing depth and lake size faster
than their importance in providing structured habitats (Fig. 29.1). This implies
that, by providing structure, macrophytes may still playa role affecting biotic
interactions in situations in which they may have no significant effect on
water-quality, nor are they important for nutrient cycling, nor as a source of
organic matter.
A. Gasith and M.V. Hoyer
386
III
c..l
c:
n:J
l-
e
c.
E
n:J
III
a::
lake size
Figure 29.1. Comparison of the changing relative importance of limnological, metabolic
and biotic effects of macrophytes along increasing lake size gradient. A positive relation
between size (surface area) and depth is assumed.
It is reasonable to assume that a PVI exceeding 40% is required for macrophytes to be able to change the water-quality conditions of an entire lake ecosystem (Canfield and Jones, 1984); this would be a situation more typical of
marshes and shallow eutrophic lakes (Canfield and Hoyer, 1992). Large-scale
oxygen depletion, for example, is most likely to occur following rapid senescence
of dense macrophyte stands in warm, poorly circulated waters (Carpenter and
Greenlee, 1981). Indeed, the fish community of densely vegetated wetlands is
composed of the most tolerant species that are able to function in dense vegetation
and survive periodic high temperatures and low dissolved oxygen levels (MacCrimmon, 1980; Johnson, 1989), The effects of macrophytes on sediment and
water quality (Carpenter and Greenlee, 1981; Carpenter and Lodge, 1986) are
expected to be restricted to the plant bed, particularly in large and deep water
bodies. It is possible, however, that biotic changes in the littoral zone in response
to chemical-physical gradients (e.g., change in composition and abundance of
prey organisms) will be carried across habitat boundary and influence lirnnetic
communities using littoral resources.
Organic matter originating from the littoral zone may have metabolic importance especially in small shallow lakes where macrophytes are highly productive
(Sculthrope, 1967; Wetzel and Hough 1973; Wetzel, 1979; Carpenter, 1983; Carpenter and Lodge, 1986). In large and deep lakes, the proportion occupied by the
littoral region is often less than 10% of the total lake area (Gasith, 1991). In the
Great Lakes, for example, the important spawning and nursery areas of most fish
species are in littoral water less than 10 m deep (Goodyear et al., 1982; O'Gorman,
1983). In addition, large lakes are often more oligotrophic and support a lesser
growth of aquatic macrophytes. In these and in highly eutrophic lakes, phytoplankton may dominate the production of organic matter and control the recycling
of nutrients (Carpenter, 1983; Hough et aI., 1989; Tilzer, 1990).
The effect of decreasing plant abundance with increasing lake size on biotic
interactions is unclear. Biotic interactions can be influenced over a wide range of
29. The Importance of Lake Size and Depth
387
plant abundance. In the absence of alternative sources for physical structure, even
sparse vegetation or isolated patches of macrophyte beds can be important in
providing substrate for colonization, refuge, feeding, and spawning grounds. For
example, in Orange Lake (Florida) young bluegills were found primarily in small
isolated islands of panic grass (Panicum spp.), which constituted less than 2% of
the lake's area (Conrow et al., 1990). B1uegills have been reported to prefer lateral
concealment (Casterlin and Reynolds, 1978) and probably favored panic grass,
which provided both protection from predators as well as access to open-water
zooplankton (Conrow et aI., 1990). In another case, Danehy (1984) and Danehy et
ai. (1991) found greater diversity and abundance of fish at relatively isolated
cobbles and rubble sites than at sandy sites in Lake Ontario. Moreover, Danehy et
ai. (1991) found that yellow perch captured at the structured sites grew faster than
those collected from the sandy sites. They attributed this difference in growth to
lower energy expenditure associated with greater cover and lower predation risk
as well as to higher food availability at the structured sites. At the sandy sites,
individuals may have been required to "commute" more in search of cover and
food. This led them to conclude that even small structured habitats may be
important to local fish populations. The significance that this may have in the
context of the whole lake ecosystem is yet to be evaluated. Another example in
which a relatively limited plant structure can be important in a large lake situation
is illustrated by the evidence that although spawning on macrophytes is unusual
for salmonids, at least a portion of the population of lake trout (Salvelinus namaycush) in Lake Tahoe spawns in deepwater mounds (40--60 m deep) over beds
of Chara (Beauchamp et al., 1992). No evidence of spawning was found over
rocky formations that exist at various depths in the lake. Apparently, the Chara
mounds are favored as they provide the basic requirements for successful egg
incubation by anchoring the eggs against currents and providing protection from
effective invertebrate and small vertebrate egg predators (Beauchamp et aI., 1992).
Similarly, it has been suggested that macrophyte beds provide cover for predationvulnerable grazers such as large herbivorous zooplankton (Timms and Moss,
1984; Davies, 1985; Jeppesen et aI., 1991; Moss et al., 1994; Lauridsen and
Lodge, 1996; Jeppesen et al., this volume, Chapter 5; Lauridsen et aI., this volume,
Chapter 13). Survival of herbivorous zooplankton even in limited macrophyte coverage may accelerate establishment of larger popUlations (Lauridsen et al., 1996) that
may, in turn, playa role in the switch from algae dominance to macrophytes (Scheffer
et al., 1993; Hargeby et al., 1994; Jeppesen et al., this volume, Chapter 28). Due to
their limited capacity for horizontal movement relative to fish, zooplankton would
probably benefit less from scattered isolated plant beds than would fish. Restricted
plant cover may therefore be expected to provide more effective refuge for zooplankton populations in small rather than large lake situations. Fish, however, are
probably able to exploit isolated plant beds over a wider lake size range.
Freshwater fish use vegetation for cover (Crowder and Cooper, 1982; Tabor
and Wurtsbaugh, 1991), foraging on benthos, epifauna, and prey organisms in the
water among the vegetation (Fairchild, 1982; Mittelbach, 1984; Heck and Crowder, 1991; Diehl and Komij6w, this volume, Chapter 2) directly as food (Prejs,
388
A. Gasith and M.V. Hoyer
1984) and as spawning and nursery sites (Goodyear et aI., 1982; O'Gorman, 1983;
Beauchamp et aI., 1992). Most of the information on the use of structured habitats
by fish is based on daytime studies. There is evidence, however, of much higher
fish density in littoral habitats at night (Beauchamp et aI., 1994) as well as a
difference in size distribution of the fish between the day and night-time littoral
zone assemblages (Gasith, Gafny, and Goren, unpublished data, Lake Kinneret).
Further studies are needed to assess the importance of diurnal shifts in fish
abundance, size, and species composition of littoral habitats.
As lake size and depth increase, macrophyte abundance declines, and structured habitats and associated resources may become in short supply (Gasith, 1991;
Beauchamp et al., 1994). Consequently, competition over littoral resources (Mittelbach, 1988), particularly among species moving from the lirnnetic zone into the
littoral region, is expected to increase with increasing lake size and depth (Gasith
and Gafny, this volume, Chapter 24). In addition, unlike abiotic structures (e.g.,
rocky formations) macrophytes undergo temporal and spatial variations. In lakes
where physical structure is provided mostly by macrophytes, organisms using
littoral resources are forced to synchronize with the "window of opportunity"
provided by macrophyte growth. If this is indeed so, competitive interactions over
macrophyte-supported resources should be highest in large deep lakes where the
abundance of macrophytes is low and in lakes where the period of macrophyte
growth is shortest (e.g., high latitudes).
Conclusion
The changing influence of macrophytes along lake size and depth gradients is
currently mostly speculative. Generally, the importance of macrophytes is expected to be proportional to their abundance in the water body, and thus their
influence will decline with increasing lake size and depth. Existing information
suggests that macrophytes can affect biotic interactions in situations in which they
have no more limnological or metabolic significance. We therefore may conclude
that only in shallow and small lakes can macrophytes potentially have significant
effects on the physical-chemical condition in the water and sediment, on internal
nutrient loading, and on lake productivity as well as on biotic interactions. In large
deep lakes, macrophyte influence on lake ecosystem diminishes and is probably
limited to some effect on biotic interactions.
Relatively small and isolated plant beds may have greater importance than have
so far been assumed. In this connection, it is possible that cases of unexplained
changes in zooplankton community structure and in fish popUlation size and
juvenile growth rate were linked to overlooked changes in the availability of
structured habitats in the littoral zone.
A better understanding of macrophyte importance in relation to lake morphometry may require separate assessment of macrophyte effects on lirnnological
conditions, metabolic processes, and biotic interactions. Due to the experimental
limitations of ecosystem manipUlation, particularly of large lakes, this will probably
29. The Importance of Lake Size and Depth
389
be achieved by long-term and comparative studies and possibly by more extensive
use of artificial structures in lakes of various sizes.
Acknowledgments. The assistance of Merav Bing and Susan Gilman of the Institute for Nature Conservation Research, of Naomi Paz of the Zoology Department, Tel-Aviv University, and of the staff of the Department of Fisheries and
Aquatic Sciences, University of Florida, Gainesville, is gratefully acknowledged.
We thank John Barko, Sarig Gafny, and Daniel E. Canfield for constructive
remarks.
References
Barko, J.W.; Smart, R.M. Effects of organic matter additions to sediments on the growth of
aquatic plants. J. Ecol. 71:161-175; 1983.
Beauchamp, D.A.; Allen, B.B.; Richards R.c.; Wurtsbauch, w.A.; Goldman, c.R. Lake
trout spawning in deepwater macrophyte beds. North Am. J. Fish. Manage. 12:442-449;
1992.
Beauchamp, D.A.; Byron, E.R.; Wurtsbauch, W.A. Summer habitat use by littoral-zone
fishes in Lake Tahoe and effects of shoreline structures. North Am. J. Fish. Manage.
14:385-394; 1994.
Canfield, D.E., Jr.; Bachmann, R.W. Predictions oftotal phosphorus concentrations, chlorophyll-a and Secchi depth in natural and artificial lakes. Can. J. Fish. Aquat. Sci.
38:414-423; 1981.
Canfield, D.E., Jr.; Hoyer, M.V. Aquatic macrophytes and their relation to the limnology of
Florida lakes. Final Report. Bureau of Aquatic Plants Management, Florida Department
of Natural Resources, Tallahassee, FL; 1992.
Canfield, D.E., Jr.; Jones, R.J. Assessing the trophic status of lakes with aquatic macrophytes. Lake and reservoir management. EPA 440/5-84-001. Proceedings of the 3rd
Annual Conference, Knoxville, TN, 1984.
Canfield, D.E., Jr.; Langeland, K.A.; Linda, S.B.; Haller, W.T. Relations between water
transparency and maximum depth of macrophyte colonization in lakes. J. Aquat. Plant
Manage. 23:25-28; 1985.
Carpenter, S.R. Enrichment of Lake Wingra, Wisconsin, by submerged macrophyte decay.
Ecology 61:1145-1155; 1980.
Carpenter, S.R. Submersed vegetation: an internal factor in lake ecosystem succession. Am.
Nat. 118:372-389; 1981.
Carpenter, S.R. Submersed macrophyte community structure and internal loading: relationship to lake ecosystem productivity and succession. In: Taggart, J., ed. Lake restoration,
protection and management. Washington, DC: U.S.E.P.A.; 1983: 105-111.
Carpenter, S.R.; Gasith, A. Mechanical cutting of submersed macrophytes: immediate
effects on littoral water chemistry and metabolism. Wat. Res. 12:55-57; 1978.
Carpenter, S.R.; Greenlee, J.K. Lake deoxygenation after herbicide use: a simulation model
analysis. Aquat. Bot. II: 173-186; 1981.
Carpenter, S.R.; Lodge, D.M. Effects of submersed macrophytes on ecosystem processes.
Aquat. Bot. 26:341-370; 1986.
Casterlin, M.E.; Reynolds, W.W. Habitat selection by juvenile bluegill sunfish, Lepomis
macrochirus. Hydrobiologia 59:75-79; 1978.
Cattaneo, L.B. Grazing on epiphytes. Limnol Oceanogr. 28:124-132: 1983.
Cattaneo, L.B.; Kalff, J. The relative contribution of aquatic macrophytes and their epiphytes to the production of macrophyte beds. Umno!. Oceanogr. 25:280-289; 1980.
Chambers, P.A. Nearshore occurrence of submerged aquatic macrophytes in relation to
wave action. Can. J. Fish. Aquat. Sci. 44:1666-1669; 1987.
390
A. Gasith and M.Y. Hoyer
Chambers, P.A.; Kalff, J. Depth distribution and biomass of submerged macrophyte communities in relation to Secchi depth. Can. J. Fish. Aquat. Sci. 42:701-709; 1985.
Chick, J.H.; Mcivor, c.c. Patterns in the abundance and composition of fishes among beds
of different macrophytes: viewing a littoral as a landscape. Can. J. Fish. Aquat. Sci.
51:2873-2882; 1994.
Conrow, R.; Zale, A.V.; Gregory, RW. Distribution and abundance of early life stages of
fishes in a Florida lake dominated by macrophytes. Trans. Am. Fish. Soc. 119:521-528;
1990.
Coops, H.; Boeters, R; Smit, H. Direct and indirect effects of wave attack on helophytes.
Aquat. Bot. 41:333-352; 1991.
Crowder, L.B.; Cooper, W.E. Habitat structural complexity and the interaction between
bluegills and their prey. Ecology 63: 1802-1813; 1982.
Danehy, RJ. Comparative ecology of fishes associated with natural cobble shoals and sand
substrates in Mexico Bay, Lake Ontario. M.S. thesis, S.U.N.Y. College of Environmental Science and Forestry, Syracuse, NY; 1984.
Danehy, RJ.; Ringler, N.H.; Gannon, J.E. Influence of nearshore structure on growth and
diets of yellow perch (Perca flavesens) and white perch (Marone americana) in Mexico
Bay, Lake Ontario. J.Great Lake Res. 17:183-193; 1991.
Davidson-Amott, RG.D.; Pollard, W.H. Wave climate and potential longshore sediment
transport, Nottawasaga Bay, Ontario. J. Great Lake Res. 6:54-67; 1980.
Davies, J. Evidence for diurnal horizontal migration in Daphnia hyalina lacustris Sars.
Hydrobiologia 120: 103-105; 1985.
Diehl, S. Foraging efficiency of three freshwater fishes: effects of structural complexity and
light. Oikos 53:207-214; 1988.
Dionne, M.; Folt, C.L. An experimental analysis of macrophyte growth forms as fish
foraging habitat. Can. J. Fish. Aquat. Sci. 48:123-131; 1991.
Duarte, C.M.; Kalff, J. Littoral slope as a predictor of the maximum biomass of submerged
macrophyte communities. Limnol. Oceanogr. 3: 1072-1080; 1986.
Duarte, C.M.; Kalff, J. Latitudinal influence on the depth of maximum colonization and
maximum biomass of submerged angiosperms in lakes. Can. J. Fish. Aquat. Sci. 44:
1759-1764; 1987.
Duarte, C.M.; Kalff, J.; Peters, R.H. Patterns in biomass and cover of aquatic macrophytes
in lakes. Can. J. Fish. Aquat. Sci. 43:1900-1908; 1986.
Dvorak, J.; Best, E.P.H. Macro-invertebrates communities associated with macrophytes of
Lake Vechten: structural and functional relationships. Hydrobiologia 95: 115-126; 1982.
Eadie, J.; Keast, A. Resource heterogeneity and fish species diversity in lakes. Can. J. Zool.
62:1689-1695; 1984.
Emery, A.R The basis of fish community structure: marine and freshwater comparison.
Environ. BioI. Fish. 3:33-47; 1978.
Engel, S. Role and interactions of submerged macrophytes in a shallow Wisconsin lake. J.
Freshwat. Ecol. 4:329-341; 1988.
Fairchild, G.W. Population responses of plant-associated invertebrates to foraging by largemouth bass fry (Micrpterus salmanides). Hydrobiologia 96:169-176; 1982.
Gasith, A. Can littoral resources influence ecosystem processes in large, deep lakes? Verh.
Int. Verein. Limnol. 24:1073-1076; 1991.
Godshalk, G.L.; Wetzel, RG. Decomposition of aquatic angiosperms. Aquat. Bot. 5:281354; 1978.
Goodyear, C.D.; Edsall, T.A.; Ormsby Dempsy, D.H.; Moss, G.D.; Polanski, P.E. Atlas of
the spawning and nursery areas of Great Lakes fishes. Fish and Wildlife Service, U.S.
Department of the Interior, Washington, DC; 1982.
Guillory, Y.; Jones, M.D.; Rebel, M. A comparison of fish communities in vegetated and
beach habitats. Florida Sci. 42: 11 3-122; 1979.
Hakanson, L. The influence of wind, fetch and water depth on the distribution of sediments
in Lake Vanern, Sweden. Can. J. Earth Sci. 14:397-412; 1977.
29. The Importance of Lake Size and Depth
391
Hakanson, L. Bottom dynamics in lakes. Hydrobiologia 81 :47-57; 1981.
Hargeby, A.; Andersson, G.; Blindow, I.; Johansson, S. Trophic web structure in a shallow,
eutrophic lake during dominance shift from phytoplankton to submersed macrophytes.
Hydrobiologia 279/280:83-90; 1994.
Heck, K.L.; Crowder, L.B. Habitat structure and predator-prey interactions in vegetated
aquatic systems. In: Bell, S.S.; McCoy, E.D.; Mushinsky, H.R., eds. Habitat structure,
the physical arrangement of objects in space. London: Chapman & Hall; 1991:281-299.
Hough, R.A.; Fornwall, M.D.; Thompson, R.L.; Putt, D.A. Plant community dynamics in a
chain of lakes: principle factors in the decline of rooted macrophytes with eutrophication. Hydrobiologia 173:199-217; 1989.
Hutchinson, G.E. A treatise on limnology. Vol. 3. Limnological botany. New York: Wiley; 1975.
Jaynes, M.L.; Carpenter, S.R. Effects of vascular and nonvascular macrophytes on sediment
redox and solute dynamics. Ecology 67:875-882; 1986.
Jeppesen, E.; Kristensen, P.; Jensen, J.P.; S0ndergaard, M.; Mortensen, E.; Lauridsen, T.
Recovery resilience following a reduction in external phosphorus loading of shallow,
eutrophic Danish Lakes: duration, regulating factors and methods for overcoming resilience. Mem. 1st. Ital. Idrobiol. 48:127-148; 1991.
Johnson, D.L. Lake Erie wetlands: fisheries considerations. In: Krieger, K.A., ed. Lake Erie
estuarine systems: issues, resources, status, and management. NOAA Estuarine Programs Office, Washington, DC, NOAA EMS Series 14; 1989:257-274.
Keddy, P.A. Quantifying within-lake gradient of wave energy: interrelationships of wave
energy substrate particle size and shoreline plants in Axe Lake Ontario. Aquat. Bot.
14:41-58; 1982.
Keddy, P.A. Shoreline vegetation in Axe Lake, Ontario: effect of exposure on zonation
patterns. Ecology 62:331-344; 1983.
Kristensen, P.; S0ndergaard, M.; Jeppesen, E. Resuspension in a shallow eutrophic lake.
Hydrobiologia 228: 101-109; 1992.
Lauridsen, T.L.; Lodge, D.M. Avoidance by Daphnia magna of fish and macrophytes:
chemical cues and predator mediated use of macrophyte habitat. Limnol. Oceanogr.
41:794-798; 1996.
Lauridsen, T.L.; Pedersen, L.J.; Jeppesen, E.; S0ndergaard, M. The importance of macrophyte bed size for cladoceran composition and horizontal migration in a shallow lake. J.
Plankton Res. 18:2283-2294; 1996.
Lillie, R.A.; Budd, J. Habitat architecture of Myriophyllum spicatum L. as an index to
habitat quality for fish and macroinvertebrates. J. Freshwat. Eco1. 7: 113-125; 1992
Lorang, M.S.; Stanford, J.A. Variability of shoreline erosion and accretion within a beach
compartment of Flathead, Montana. Limnol. Oceanogr. 38: 1783--1795; 1993.
MacCrimmon, H.R. Nutrients and sediment retention in a temperate marsh ecosystem. Int.
Rev. Ges. Hydrobiol. 65:719-744; 1980.
Miller, A.c.; Beckett, D.C.; Way, C.M.; Bacon, E.J. The habitat value of aquatic macropytes
for macroinvertebrates. Technical Report A-89-3, U.S. Army Corps of Engineers, Washington, DC; 1989.
Mittelbach, G.G. Predation and resource partitioning in two sunfishes (Centrachidae).
Ecology 65:499-513; 1984.
Mittelbach, G.G. Competition among refuging sunfishes and effects of fish density on
littoral zone invertebrates. Ecology 69:614-623; 1988.
Morin, J.O. Initial colonization of periphyton on natural and artificial apices of Myriophyllum heterophyllum Michx. Freshwat. BioI. 16:685--694; 1986.
Moss, B. The micro waterscape-a four dimensional view of the inkractions among water
chemistry, phytoplankton, periphyton, macrophytes, animals and ourseleves. Wat. Sci.
Techn.32:105-116; 1995.
Moss, B.; McGowan, S.; Carvalho, L. Determination of phytoplankton crop by top-down
and bottom-up mechanisms in a group of English lakes, the West Midland meres.
Limnol. Oceanogr. 39: 1020-1029; 1994.
392
A. Gasith and M.Y. Hoyer
O'Gonnan, R. Distribution and abundance of larval fish in the nearshore waters of western
Lake Erie. 1. Great Lakes Res. 9:14-22; 1983.
Osgood, R.A. Lake mixis and internal phosphorus dynamics. Arch. Hydrobiol. 113:629638; 1988.
Pearsall, W.H. The aquatic and marsh vegetation of Esthwaite water. J. Ecol. 5:180-201;
1917.
Pearsall, W.H. The aquatic vegetation ofthe English lakes. J. Ecol. 8:163-201; 1920.
Pearsall, W.H. Dynamic factors affecting aquatic vegetation. Proc. Int. Congr. Plant Sci.
1:667-672; 1929.
Planter, M. Physical and chemical conditions in the helophyte zone of the lake littoral. Pol.
Arch. Hydrobiol. 20: 1-7; 1973.
Prejs, A. Herbivory by temperate freshwater fishes and its consequences. Environ. BioI.
Fish. 10:281-296; 1984.
Prentki, R.T.; Adams, M.S.; Carpenter, S.R.; Gasith, A.; Smith, C.S.; Weiler, P.R. The role
of submerged weedbeds in internal loading and interception of allochthonous materials
in Lake Wingra, Wisconsin. Arch. Hydrobiol. Suppl. 57(2):221-250; 1979.
Rounsefell, G.A. Fish production in lakes as a guide for estimating production in proposed
reservoirs. Copeia 1:29-40; 1946.
Savino, J.E; Stein, R.A. Predator-prey interaction between largemouth bass and bluegills
as influenced by simulated submerged vegetation. Trans. Am. Fish. Soc. 111:255-266;
1982.
Scheffer, M.; Hosper, S.H.; Meijer, M.L.; Moss, B.; Jeppesen, E. Alternative equilibria in
shallow lakes. Trends. Ecol. Evol. 8:275-279; 1993.
Schram, H.L., Jr.; Jirka, K.J.; Hoyer, M.Y. Epiphytic macroinvertebrates on dominant
macrophytes in two central Florida lakes. 1. Freshwat. Ecol. 4: 151-161; 1987.
Schriver, P.; Bl'lgestrand, J.; Jeppesen, E.; Sl'lndergaard, M. Impact of submerged macrophytes on the interactions between fish, zooplankton and phytoplankton: large scale
enclosure experiments in shallow eutrophic lake. Freshwat. BioI. 33:255-270; 1995.
Sculthrope, C.D. The biology of aquatic vascular plants. London: Edward Arnold; 1967.
Sozska, GJ. Ecological relations between invertebrates and submerged macrophytes in the
lake littoral. Ekol. Pol. 23:393-415; 1975.
Spence, D.H.N. The zonation of plants in freshwater lakes. In: MacFadyan, A.; Ford, E.D.,
eds. Advances in ecological research. London: Academic Press; 1982:37-126.
Tabor, R.A.; Wurtsbaugh, W.A. Predation risk and the importance of cover for juvenile
rainbow trout in lentic systems. Trans. Am. Fish. Soc. 120:728-738; 1991.
Tilzer, M.M. Specific properties of large lakes. In: Tilzer, M.M.; Cerruya, C., eds. Large
lakes, ecological structure and function. New York: Springer-Verlag; 1990:39-45.
Timms, R.M.; Moss, B. Prevention of growth of potentially dense phytoplankton populations by zooplankton grazing, in the presence of zooplanktivorous fish, in a shallow
wetland ecosystem. Limnol. Oceanogr. 29:472-486; 1984.
Vollenweider, R.A. Input-output models with special reference to phosphorus loading
concept in limnology. Schweiz. Z. Hydro!. 37:53-84; 1975.
Wetzel, R.G. The role of the littoral zone and detritus in lake metabolism. Arch. Hydrobiol.
13:145-161; 1979.
Wetzel, R.G. Detritus, macrophytes and nutrient cycling in lakes. Mem. 1st. Ital. Idrobiol.
47:233-249; 1990.
Wetzel, R.G.; Hough, R.A. Productivity and the role of aquatic macrophytes in lakes. An
assessment. Pol. Arch. Hydrobiol. 20:9-19; 1973.
30. Nutrient-Loading Gradient in Shallow Lakes:
Report of the Group Discussion
Stephen R. Carpenter, Ellen van Donk, and Robert G. Wetzel
Nutrient gradients are an important means of organizing lirnnological information.
Nutrients are among the most important controls of lake ecosystem processes.
Other key factors such as light, macrophytes, and predation change in systematic
ways as nutrient availability increases or decreases. Consequently, the nutrient
gradient is a useful tool for explaining our understanding of lake diversity. The
nutrient gradient can also be used to make predictions about the state of lakes as
nutrient status changes.
Our discussion of the nutrient gradient is based on several assumptions. The
nutrient gradient represents a continuum of nutrient input (or loading) rates. We do
not consider transient dynamics that result from abrupt changes in nutrient input
rate. Rather, we consider average or prevalent conditions in lakes that have
experienced a given rate of nutrient input for an extended period of time: long
enough for several hydrologic flushings, and at least several generations of the
dominant organisms.
The generalizations below are based on additional conditions pertinent to the
goals of this book. The lakes are too shallow to be thermally stratified. We assume
a temperate climate; that the hydraulic residence time, dissolved organic carbon
inputs, and alkalinity remain constant; that there are no important species invasions or extirpations; and that harvest rates of the dominant fishes remain
constant.
Given these assumptions, discussants converged on a general view of shallow
lake ecosystems across the nutrient loading gradient (Fig. 30.1). The possibility of
393
394
S.R. Carpenter, E. van Donk, and R.O. Wetzel
MACROPHYTE-PHYTOPLANKTON
?
I--_ _.;. .:A.: :.LT"'E::.;.R.;;..N;;..A;.;.T:....;IV..=Ec...:S;.;.T.;...:A"'TE::;S=--_----!I _____________ _
FLOATING-LEAVED PLANTS
SUBMERSED FL